1154649614 Q * sid3windr Server closed connection 1154649627 J * sid3windr luser@bastard-operator.from-hell.be 1154650221 Q * Hollow Server closed connection 1154650227 J * Hollow ~hollow@2001:a60:f026::1 1154650656 Q * lilo2 Remote host closed the connection 1154650701 J * lilo2 hiddenserv@tor.noreply.org 1154652640 Q * Wonka Read error: Connection reset by peer 1154652862 J * Wonka produziert@chaos.in-kiel.de 1154653211 J * MrX ~urk@219.95.5.4 1154653387 Q * MrX 1154653867 J * MrX ~urk@219.95.5.4 1154653989 Q * Bertl_oO Server closed connection 1154653998 J * Bertl_oO herbert@IRC.13thfloor.at 1154654037 J * s0undt3ch_ uxpxjm@bl7-243-104.dsl.telepac.pt 1154654183 Q * s0undt3ch Ping timeout: 480 seconds 1154654185 N * s0undt3ch_ s0undt3ch 1154655328 J * beave ~beave@telephreak.org 1154655339 J * meandtheshell ~markus@85-124-36-219.dynamic.xdsl-line.inode.at 1154655352 M * beave howdy howdy. 1154655453 M * beave I'm trying to build (gentoo/ebuild) util-vserver-0.30.210, and I'm getting a of link problems (for example, "keep-ctx-alive.c:(.text+0x1e4): undefined reference to `socket'"). I'm using the 2.6.17.7-grsec2.1.9-vs2.0.2-rc27 kernel. Anyone run into this issue before? 1154655676 N * Bertl_oO Bertl 1154655680 M * Bertl evening folks! 1154655695 M * beave Howdy Bertl.... 1154655700 N * beave Beave 1154655743 M * Bertl Beave: unusual .. maybe some library not available? what do you get from the ./configure (as last output)? 1154655774 M * Beave yeah, thats what i figured, just having trouble figuring out which lib. let me re-check my ./configure output. 1154655972 M * Beave it does the ./configure fine (creates the Makefile, displays Features and Paths) . No real errors stand out. 1154655982 M * Beave I'm building it on another machine, let me see how that goes. 1154656285 M * Beave haha. same issue on my other gentoo box. Let me poke around about. It looks like all network related stuff ('socket', 'bind', 'recv', etc) 1154657090 M * Bertl what dietlibc version is that? 1154657191 M * Beave let me check. 1154657212 M * Beave dietlibc-0.28 1154657226 M * Beave got it to build on a old box, so I'm trying to see the differences. 1154657338 J * ntrs_ ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com 1154657421 Q * lilo2 iridium.oftc.net neutron.oftc.net 1154657421 Q * Hunger iridium.oftc.net neutron.oftc.net 1154657421 Q * ntrs__ iridium.oftc.net neutron.oftc.net 1154657421 Q * blues^ iridium.oftc.net neutron.oftc.net 1154657421 Q * FireEgl iridium.oftc.net neutron.oftc.net 1154657421 Q * sladen iridium.oftc.net neutron.oftc.net 1154657421 Q * Vudumen iridium.oftc.net neutron.oftc.net 1154657421 Q * x03 iridium.oftc.net neutron.oftc.net 1154657421 Q * click iridium.oftc.net neutron.oftc.net 1154657421 Q * MooingLemur iridium.oftc.net neutron.oftc.net 1154657421 Q * SNy iridium.oftc.net neutron.oftc.net 1154657421 Q * bubulak iridium.oftc.net neutron.oftc.net 1154657421 Q * meebey iridium.oftc.net neutron.oftc.net 1154657421 Q * h01ger iridium.oftc.net neutron.oftc.net 1154657530 J * lilo2 hiddenserv@tor.noreply.org 1154657530 J * Hunger Hunger.hu@Hunger.hu 1154657530 J * ntrs__ ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com 1154657530 J * blues^ blues@blysk.ds.pg.gda.pl 1154657530 J * FireEgl ~FireEgl@Atlantica.CJB.Net 1154657530 J * sladen paul@starsky.19inch.net 1154657530 J * Vudumen ~vudumen@217.20.138.14 1154657530 J * x03 ~x03@10001268031.0000030345.acesso.oni.pt 1154657530 J * click click@ti511110a080-4165.bb.online.no 1154657530 J * MooingLemur ~troy@shells200.pinchaser.com 1154657534 J * SNy 31b763fc31@bmx-chemnitz.de 1154657534 J * h01ger ~holger@socket.layer-acht.org 1154657534 J * bubulak ~bubulak@whisky.pendo.sk 1154657534 J * meebey meebey@booster.qnetp.net 1154657553 Q * ntrs__ Read error: Connection reset by peer 1154657644 J * Savvy ~ali@81.215.91.178 1154657666 M * Bertl welcome Savvy! 1154657863 M * Savvy hi 1154657913 M * Beave Doh - yeah, both the work and non-working box are using dietlibc 0.28 1154657967 M * Bertl maybe something with the gcc or library path? 1154657980 M * Beave maybe. 1154657998 M * Beave i think this is just gonna take some poking around. 1154658210 M * Beave any idea where sys/capability.h comes from? 1154658240 M * Bertl recent distros have a libcapability or libcap 1154658370 M * Beave thanks bertl. 1154658491 M * Bertl you're welcome! 1154658958 M * Beave src/keep-ctx-alive.o: In function `Esocket': 1154658959 M * Beave ./ensc_wrappers/wrappers-socket.hc:26: undefined reference to `socket' 1154658961 M * Beave damn, 1154658969 M * Beave let me go look'see. 1154659533 Q * insomniac Remote host closed the connection 1154659832 J * insomniac ~insomniac@slackware.it 1154661079 J * Aiken_ ~james@tooax7-040.dialup.optusnet.com.au 1154661291 J * nn ~sean@210-84-58-60.dyn.iinet.net.au 1154661339 M * nn has anyone packaged an RPM from apache 2 source? 1154661362 M * Bertl hmm, how does that relate to Linux-VServer? 1154661389 M * nn umm.. I am not sure - I am unable to find a room to ask. 1154661403 Q * Aiken Ping timeout: 480 seconds 1154661417 M * Bertl hmm, check out pbone.net or rpmfind.net 1154661427 M * nn thanks 1154661441 P * nn 1154661602 J * Skram ~MarkS@admins.sentiensystems.net 1154661653 M * cehteh morning 1154661663 M * Bertl good morning! 1154663637 M * Skram hello all 1154663652 M * Skram Beave: get it worked out? 1154664172 Q * s0undt3ch Ping timeout: 480 seconds 1154664530 M * Beave still working on it. 1154664538 M * Beave i took a break and watched tv for a bit. 1154665381 J * s0undt3ch xmmeptny@bl7-241-149.dsl.telepac.pt 1154665640 M * Bertl Savvy: hmm, testing the ML with a 'test' message? why not ask something useful :) 1154665798 M * cehteh as in "Can you read this test message?" .. :) 1154665865 M * Bertl lol 1154665881 M * cehteh Bertl: did you seen my recent question in #kernelnewbies about letting the OOM killer send a SIGDANGER to processes it wants to kill to let them reduce their working set, if they handle it? 1154665908 M * Bertl nope, but sounds interesting ... 1154665913 M * cehteh are you familar with the OOM killer, would such a change be trivial or very complicated.. 1154665947 M * cehteh i only did a fast look .. there is already code which sends a SIGTERM for processes with RAWIO access 1154665993 M * cehteh well for garbagge collrector implementors and maybe even libc it would be nice but it is just an idea so far 1154665995 M * Bertl well, the signal sending is trivial 1154666026 M * Bertl the problem is more that the OOM killer is only invoked when the is immanent need to free memory ... 1154666036 M * Bertl *there 1154666050 M * cehteh yes .. but does the OOM killer expect that memory is freed when he sends the signal, and does it kill shortly before the memory is exhausted or at the point when memory is exhausted 1154666062 M * Bertl so, it cannot simply send a signal, reschedule the app and wait that it does something 1154666067 M * cehteh exactly 1154666111 M * cehteh well it can stop the process which wants more memory as long the signal is pending and implement some timing policy until that signal has to be handled 1154666123 M * cehteh after it might be not that trivial 1154666132 M * Bertl and, if you do that in two stages (i.e. some kind of OOM warn thread, and a killer) they have to be synchronized, so that they actually signal/kill the same app 1154666142 Q * shedi Read error: Connection reset by peer 1154666149 M * cehteh anyways .. reducing the heap and free memory to the system isnt trivial for the libc either 1154666154 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1154666207 M * cehteh implementing proper getrlimit for virtoal memory and other stats might be done before that anyways 1154666251 M * cehteh ok .. well just an idea, maybe someone does that someday .. maybe me if i really need it 1154666292 M * Bertl it's always good to discuss ideas, maybe somebody has another idea based on yours ... 1154666309 M * cehteh SIGDANGER is actually the way other unixes handle that 1154666372 M * cehteh i had also the idea about disabling overcommit per process .. but thats likely harder to implement and opens a can of worms in sense of security 1154666387 J * coocoon ~coocoon@p54A066A0.dip.t-dialin.net 1154666412 M * cehteh since pages for such processes have to be granted, they can easily DoS the vm 1154666429 M * cehteh and i want a solution which doesnt require special privileges 1154666474 M * Bertl not only that, more important overcommitted memory is something which is mainly caused by fork() .. which makes the whole scenario really ugly 1154666539 M * Bertl okay, I'm off to bed now ... back tomorrow! have fun! 1154666548 N * Bertl Bertl_zZ 1154666550 M * cehteh (maybe a grant limit, which, if set can never overun) 1154666562 M * cehteh well yes .. ok n8 :) 1154667426 M * Beave yay. got it build. 1154667442 M * Beave had to rebuild dietlibc with -nopie. 1154667451 M * cehteh heh 1154667543 M * Skram Beave: awesome 1154667594 J * shedi ~siggi@inferno.lhi.is 1154667623 M * Skram if i want vpses to be able to talk to each other via hostname.. is there a better way than just adding them into /etc/hosts and/or mounting that into each VServer? 1154667688 M * Beave DNS :) 1154668390 M * cehteh NIS :) 1154668401 M * cehteh (jk) 1154668446 M * cehteh ldap 1154668769 M * Skram meh 1154669141 Q * Johan Server closed connection 1154669169 J * Johan ~finger@lounge.datux.nl 1154669202 Q * id23 Ping timeout: 480 seconds 1154669778 J * id23 ~id@p50813EEB.dip0.t-ipconnect.de 1154670259 Q * meandtheshell Quit: bye bye ... 1154671207 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1154671391 J * dna ~naucki@237-210-dsl.kielnet.net 1154671691 J * coocoon ~coocoon@p54A06EF0.dip.t-dialin.net 1154672152 Q * MooingLemur Quit: Leaving 1154673197 Q * shedi Ping timeout: 480 seconds 1154673273 M * Hollow Beave: dietlibc-0.30 ebuild would have done this for you 1154674232 Q * lilo2 Remote host closed the connection 1154674327 J * lilo2 hiddenserv@tor.noreply.org 1154674477 Q * gerrit_ Ping timeout: 480 seconds 1154674708 Q * lilo2 Quit: leaving 1154674870 J * lilo2 hiddenserv@tor.noreply.org 1154675133 J * renihs ~penguin@83-65-34-34.arsenal.xdsl-line.inode.at 1154675170 J * Viper0482 ~Viper0482@p54974DCA.dip.t-dialin.net 1154675178 M * renihs he bertl! i just noticed bugs.linux-vserver.org has an interesting touch to it :) 1154675369 Q * lilo2 Ping timeout: 480 seconds 1154675512 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1154675523 Q * Viper0482 Remote host closed the connection 1154675546 J * coocoon ~coocoon@p54A06EF0.dip.t-dialin.net 1154675661 J * lilo2 hiddenserv@tor.noreply.org 1154675827 J * ||Cobra|| ~cob@pc-csa01.science.uva.nl 1154677231 M * Skram work in less than 4 hours.. peace out all 1154677301 M * Beave Hollow: oh, thanks. I'll re-sync and use .30 next time. 1154677313 M * Beave I blew the server up, so I need to restart tomorrow. :) 1154677592 J * yarihm ~yarihm@whitehead2.nine.ch 1154677885 J * mire_ ~mire@61-166-222-85.COOL.ADSL.VLine.Verat.NET 1154678357 Q * Aiken_ Ping timeout: 480 seconds 1154679612 Q * mire_ Ping timeout: 480 seconds 1154679665 Q * phreak`` Server closed connection 1154679694 J * phreak`` ~phreak``@140.211.166.183 1154679737 Q * lilo2 Remote host closed the connection 1154679822 J * lilo2 hiddenserv@tor.noreply.org 1154680085 Q * phreak`` Quit: leaving 1154680992 Q * dna Quit: Verlassend 1154681030 Q * SNy Server closed connection 1154681031 J * SNy 6cfbac777d@bmx-chemnitz.de 1154681036 J * dna ~naucki@237-210-dsl.kielnet.net 1154681232 Q * dna 1154681249 J * dna ~naucki@237-210-dsl.kielnet.net 1154681406 J * phreak`` ~phreak``@140.211.166.183 1154681609 Q * dna Quit: Verlassend 1154681650 J * dna ~naucki@237-210-dsl.kielnet.net 1154682219 Q * renihs Quit: Leaving 1154682273 Q * dna Quit: Verlassend 1154682733 Q * michal` Ping timeout: 480 seconds 1154683043 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1154683114 J * schimmi ~sts@port-212-202-73-176.dynamic.qsc.de 1154683306 J * michal` ~michal@www.rsbac.org 1154684265 Q * doener Server closed connection 1154684291 J * doener ~doener@host.magicwars.de 1154684299 J * dna ~naucki@237-210-dsl.kielnet.net 1154684302 Q * dna 1154684316 J * dna ~naucki@237-210-dsl.kielnet.net 1154684802 Q * lilo2 Server closed connection 1154684973 Q * daniel_hozac Server closed connection 1154684984 J * daniel_hozac ~daniel@c-2c1472d5.010-230-73746f22.cust.bredbandsbolaget.se 1154684996 Q * kaner iridium.oftc.net nobelium.oftc.net 1154684996 Q * pusling iridium.oftc.net nobelium.oftc.net 1154684996 Q * lilalinux iridium.oftc.net nobelium.oftc.net 1154684996 Q * mountie iridium.oftc.net nobelium.oftc.net 1154684996 Q * Radiance iridium.oftc.net nobelium.oftc.net 1154684996 Q * waldi iridium.oftc.net nobelium.oftc.net 1154684996 Q * samuel_ iridium.oftc.net nobelium.oftc.net 1154684996 Q * dhansen iridium.oftc.net nobelium.oftc.net 1154684996 Q * Snow-Man iridium.oftc.net nobelium.oftc.net 1154684996 Q * weasel iridium.oftc.net nobelium.oftc.net 1154684996 Q * bragon iridium.oftc.net nobelium.oftc.net 1154684996 Q * ray6 iridium.oftc.net nobelium.oftc.net 1154684996 Q * derjohn iridium.oftc.net nobelium.oftc.net 1154685098 J * weasel weasel@asteria.debian.or.at 1154685146 J * kaner kaner@strace.org 1154685169 J * bragon ~weechat@sd866.sivit.org 1154685185 J * Snow-Man ~sfrost@kenobi.snowman.net 1154685188 J * pusling pusling@195.215.29.124 1154685188 J * Radiance fd91f6f017@halt.1984world.eu 1154685755 J * waldi ~waldi@83.137.100.38 1154685787 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com 1154685853 J * lilalinux ~plasma@h1-gw.of.net-lab.net 1154685880 J * derjohn ~derjohn@80.69.37.19 1154685945 J * dhansen ~dave@sprucegoose.sr71.net 1154685973 J * samuel_ ~samuel@jupe.quebectelephone.com 1154686874 Q * Hunger Server closed connection 1154686879 J * Hunger Hunger.hu@Hunger.hu 1154688892 P * anonc adios 1154689485 J * lilo2 ~0710AAD4@149.9.0.27 1154690190 J * shedi ~siggi@213.190.108.50 1154690735 Q * blues^ Server closed connection 1154690738 J * blues blues@blysk.ds.pg.gda.pl 1154690998 J * pisco ~pampel@p5087B3C8.dip0.t-ipconnect.de 1154691626 Q * FireEgl Server closed connection 1154692183 Q * insomniac Ping timeout: 480 seconds 1154692257 J * FireEgl ~FireEgl@2001:5c0:84dc:: 1154692494 Q * yarihm Quit: Leaving 1154692685 J * ray6 ~ray@vh5.gcsc2.ray.net 1154693686 Q * abi Server closed connection 1154693687 J * abi ~abi@enz.schiach.de 1154693831 Q * lilalinux Remote host closed the connection 1154693876 J * lilalinux ~plasma@h1-gw.of.net-lab.net 1154693981 M * Hollow doener, daniel_hozac: do you guys have a small pic for the new wiki? google images did not find any :) 1154694009 M * doener of myself? 1154694135 M * doener I doubt that... at least none I'd like to see on the net ;) 1154694282 M * Hollow heh 1154694296 M * Hollow doener: http://vwiki.croup.de/Developers 1154694335 M * bubulak lol 1154694336 M * bubulak :) 1154694347 M * Hollow hm? :) 1154694357 M * bubulak nice photos 1154694368 M * Hollow yeah, i look so awesome.. *ahem* 1154694376 M * bubulak :)) 1154694394 M * dna the whole new site looks very nice 1154694397 M * bubulak no just hebert looks like from IT Crowd :) 1154694414 M * Hollow dna: thanks :) 1154694417 M * Hollow bubulak: hehehe 1154694729 J * _Kara ~Kashira@ip-80-226-199-202.vodafone-net.de 1154694735 M * _Kara mornin 1154695012 J * meandtheshell ~markus@85-124-36-41.dynamic.xdsl-line.inode.at 1154695127 M * doener Hollow: well, if I now could find my usb cable for my cam... 1154695271 M * Hollow :) 1154695324 M * sid3windr :) 1154695346 M * doener http://www.13thfloor.at/~doener/fresh_showered.jpg -- only 5 minutes old ;) 1154695405 M * doener hm, a little huge... well, you can cut it down to a reasonable size 1154695489 M * _Kara you look kinda - overslept :P 1154695516 M * sid3windr doener: you look awake ;) 1154695516 M * doener mission accomplished then :) 1154695713 M * Hollow heh 1154695714 M * Hollow thanks 1154696038 M * yang what is wrong with this command? ./chbind6 -n vserver2 -a -6 2a01:b0:1114::25:33/64 1154696041 M * yang IPv6: 2a01:b0:1114::25:33/64 1154696041 M * yang chbind6: vc_net_create(): Invalid argument 1154696182 M * Hollow doener: now everyone knows you ;) 1154696241 M * sid3windr ohnoes 1154696483 M * doener Hollow: good, I'll remove the big one then 1154696744 M * yang To get the syscall number, check the syscall# line in vserver-info's output. which vserver-info output ? 1154696995 M * Hollow yang: the one if you call vserver-info 1154697120 M * yang ok 1154697394 Q * schimmi Ping timeout: 480 seconds 1154697507 M * yang it doesn't work, it doesn't display IPv6 address in guest 1154697566 Q * sladen Server closed connection 1154697572 J * sladen paul@starsky.19inch.net 1154697759 N * Bertl_zZ Bertl 1154697764 M * Bertl morning folks! 1154697788 M * doener morning Bertl 1154697789 M * Bertl bubulak: did you try turning it off an on again? 1154697800 M * bubulak lool 1154697836 M * doener Bertl: just in time ;) was there any IPC bug fixed since 2.1.0-rc7? (if you happen to remember ;) 1154697886 M * Bertl 2.1.0-rc7 hmm, that's quite a long time ... 1154697893 M * doener indeed :) 1154697927 M * Bertl in linux-vserver code or mainline, or both? 1154697947 M * doener linux-vserver, I'd opt for missing isolation 1154697990 M * Hollow morning Bertl! while searching for the database schema of the current wiki (your dump did not include it) i found out that our wiki has a list-all-pages feature :o 1154698073 M * Hollow and it revealed the whole mess.. i think we need to sort out which old pages will get a redirect, because it's just a mess otherwise 1154698128 M * Bertl I would say we add a redirect for each one, _but_ we can redirect to 'new' pages too 1154698141 M * Hollow well, that was the idea anyway 1154698155 M * Hollow like: http://vwiki.croup.de/Linux-VServer-Paper-03 1154698156 M * Bertl _and_ we only redirect as long as there _is_ no new page with that url/name 1154698189 M * Hollow but mediawiki does not support external redirection 1154698195 M * doener hm, at least semaphore isolation seems to be there in that release 1154698213 M * Bertl Hollow: hmm? 1154698231 M * Hollow Bertl: you cannot redirect from a wiki page to an external url 1154698231 J * pisc1 ~pampel@p5087BD48.dip0.t-ipconnect.de 1154698244 Q * ||Cobra|| Remote host closed the connection 1154698292 M * Bertl Hollow: well, then we have to do it at the apache level ... 1154698320 M * Bertl Hollow: what _is_ the page/url http://vwiki.croup.de/Linux-VServer-Paper-03 did redirect to? 1154698325 M * Hollow sounds like mod_rewrite fun :) 1154698337 M * Hollow http://vwiki.croup.de/Paper 1154698350 M * Hollow all 16 pages get redirected to that one 1154698357 M * Bertl hmm, why doesn't that show up somewhere? 1154698370 M * Hollow it says redirected below the headline 1154698385 M * Bertl i.e. if I have a bookmark, it will not be updated, and I see no chance to actually update it ... 1154698404 M * Bertl ... as I do not know _where_ I was 'redirectd' 1154698432 M * Hollow yeah, mediawiki does not do http redirects 1154698441 M * Bertl a 'moved permanently' header would be a better solution IMHO 1154698488 M * Bertl but as I said, we can do that via apache .. already doing that for many pages on 13thfloor 1154698523 M * Hollow probably best to make a rewrite map for all pages not copied 1154698533 M * Hollow iirc mod_rewrite can handle such maps.. 1154698568 M * Bertl yep, no problem with that ... 1154698622 Q * pisco Ping timeout: 480 seconds 1154698625 M * Hollow but i guess we can do nothing about the internal redirects without modifing mediawikis source (not a good idea) 1154698722 M * Roey hi all! 1154698725 M * Roey Hollow, Bertl! 1154698748 M * Hollow Roey! 1154698756 M * Roey Something I like about VMWare is that all the config settings for a particular vserver reside in the guest's .vmk file. 1154698771 M * Bertl http://www.heise.de/newsticker/meldung/76389 1154698783 M * Roey Now, with VServer, guest-specific files are spread out in multiple directories 1154698797 M * Roey (not counting the guest's own filesystem files) 1154698806 M * Roey is there a way you guys can make this easier for us users 1154698806 M * Roey ? 1154698808 M * Roey like 1154698814 M * Bertl Roey: if you really want that, you can put both the config and the guest into a loop device 1154698823 M * Roey but then I hsave to mount it 1154698830 M * Roey what if I dont' want to go to that hassle 1154698830 M * Roey ? 1154698832 M * Roey i.e. 1154698837 M * Bertl makes the guest a little slower (as with VMware) but you can handle that with a simple script 1154698845 M * Roey I just want one 8 GB file to hold allof the guest stuff. 1154698846 M * Hollow Bertl: we seriously need PR too ;) 1154698854 M * Roey Hollow: fix what I said above 1154698858 M * yang bonbons: awake? 1154698860 M * Roey Hollow: make it easier to allow openvpn 1154698874 M * Roey Hollow: and that will make vserver much more usable. 1154698875 M * bonbons yang: yes 1154698879 M * Roey er 1154698885 M * Roey that will make vserver that much more appealing. 1154698897 M * Hollow Roey: well, that's a matter of personal preference 1154698904 M * Bertl Hollow: somehow I came to the conclusion that every minute spent on PR is a lost one ... 1154698917 M * _Kara Bertl: Why? 1154698919 M * Roey Bertl: why? 1154698921 M * Hollow hehe 1154698928 M * yang bonbons: i compiled chbind6 with option gcc -lvserver -I /usr/src/linux-2.6.17.6/include/ -D__NR_vserver=273 -o chbind6 chbind6.c, but I still don't see the ipv6 address in my guest? 1154698929 M * bonbons was playing with chbind6, but need to check exactly whatit does... has special behavior and compilation trouble because of disabled dynamic contexts 1154698937 M * doener Roey: http://linux-vserver.org/ConfigConverters -- you search for such a thing? 1154698945 M * Bertl Hollow: look at Roey, he wants to get something which is already there, although it is inferior to Linux-VServer ... why? 1154698953 M * Roey Hollow: I find it easier to handle guests if all their specific files are packaged together instead of spread out 1154698962 M * Roey doener: haven't searched for it. 1154698973 M * Roey Bertl: what do you mean? 1154698978 M * Hollow Bertl: maybe he doesn't like so much files and symlinks ;) 1154698990 M * Hollow (as do i) 1154698991 M * bonbons yang: try with context-id instead of context name 1154698991 M * Roey Hollow: I don't like those symlinks at all. 1154699007 M * yang bonbons: yes i do that 1154699009 M * Hollow that's why vcd has all configuration in a sqlite database 1154699011 M * Roey Hollow: I think it's easy to mess up (and consequently easy to mess up the operation of the guest) 1154699030 M * yang bonbons: did 1154699038 M * Roey at this point I think I'm going to go with vmware simply because the administration would be easier. 1154699046 M * Hollow lol 1154699061 M * Hollow vmware meets totally different requirements as vserver 1154699064 M * Roey right. 1154699068 M * Bertl Roey: please do that .. we are _not_ doing VMware :) 1154699073 M * Roey Bertl: :)( 1154699100 M * yang bonbons: ipv6 on main works just fine 1154699103 M * Hollow have fun running 20 vmware images at the same time *giggle* 1154699104 M * Bertl Roey: if Linux-VServer is not what you prefer, you have exactly three options 1154699131 M * Bertl Roey: 1) improve it yourself (you have all the code/source) 1154699163 M * Bertl Roey: 2) pay or find somebody to do it for you if you can't 1154699167 M * bonbons yang: trying to get chbind6 working on my box... works with my version of vserver-utils, so chbind6 should be able to work as well... 1154699183 M * Bertl Roey: 3) make a different choice in what you use/want 1154699235 M * yang bonbons: i do have an older package of vserver-utils 0.30.204-5sarge, which version must be installed in order for IPv6/guest to work? 1154699236 M * Bertl Roey: of course, 2 includes convincing some of the developers (with arguments) that this is the right way to go ... 1154699258 M * Hollow Bertl: maybe it would be even better to redirect _all_ old pages via mod_rewrite and not via internal wiki redirect if the page is available with another name, this way we could get HTTP 302 everywhere 1154699275 M * Bertl Roey: putting guest files into a loopback file has _no_ value IMHO, only overhead and drawbacks ... 1154699279 M * bonbons yang: should not, maybe daniel_hozac has some hint here? 1154699305 M * Bertl bonbons, yang: he is on vacation - hard to believe but true :) 1154699325 M * Bertl Hollow: yes, I think that is better ... 1154699337 M * bonbons Bertl: ok, then let's see inside chbind6 in depth... 1154699346 M * Hollow Bertl: good, i will prepare the map then.. 1154699368 M * yang bonbons: make-kpkg clean doesn't reset the patches, once they were patched into the kernel tree? 1154699399 M * Hollow Bertl: btw, you asked if we can just copy the vserver image for the wiki, but the guest is gentoo of course, probably not what you want ;) 1154699435 M * Hollow i.e. should i really setup everything inside this guest, or do we move the htdocs/sql to another guest? 1154699443 M * yang bonbons: i think i did all according to the manual there 1154699456 M * yang bonbons: also i dont get any errors in dmesg 1154699488 M * Bertl Hollow: well, actually I do not care about the distro, you probably have to help maintaining the guest for some time though ... 1154699491 M * bonbons yang: right now I'm trying to understand the way daniel is doing things in chbind6 ... 1154699501 M * yang bonbons: ok take your time 1154699508 M * Hollow Bertl: sure, i can maintain it as long as needed 1154699542 M * id23 aloha Bertl :) 1154699551 M * Bertl hey id23! LTNS 1154699562 M * id23 pretty busy the last weeks 1154699586 M * id23 i see there is a new rc27 i did not compile yet ;) 1154699588 M * Hollow i.e. i will also setup the ftp/http archives, subversion and probbaly git, as well as the pastebin 1154699631 M * Hollow Bertl: and btw.. the wiki.l-v.org domain is only known by ns1.l-v.org 1154699637 M * Bertl try to keep it modular ... 1154699661 M * id23 are you currently planing to set up a new WIKI vor vServer ? 1154699671 M * Hollow id23: yes, http://vwiki.croup.de 1154699675 M * id23 nice :) 1154699689 M * id23 i will just take a look @ it 1154699742 M * id23 hey looks clean & nice 1154699753 M * Hollow thanks :) 1154699791 M * Hollow the "About" section is already copied completely 1154699827 M * Hollow i also sorted the current wiki if anyone is interested: http://home.xnull.de/misc/lv.org/ 1154699851 Q * Vudumen Server closed connection 1154699857 J * Vudumen ~vudumen@217.20.138.14 1154700314 Q * h01ger Server closed connection 1154700378 J * h01ger ~holger@socket.layer-acht.org 1154700474 M * bonbons yang: with shiny syscall it works correctly for me, ah IPv6 addresses show up in /proc/virtnet//info as expected... no invalid error. 1154700561 M * bonbons anhour ago: what is wrong with this command? ./chbind6 -n vserver2 -a -6 2a01:b0:1114::25:33/64 1154700575 M * bonbons in that command you used the guest name instead of id... 1154700733 M * yang yes, later i added context-id 1154700811 M * Hollow Bertl: what do you think about adding all pages to redirect map for the beginning, replace the wiki and remove the redirects once the page has been added to the new wiki? this way we could start the wiki next week probably although not all pages were copied.. 1154700823 M * bonbons yang: and still the same error? 1154700830 M * yang bonbons: also my virnet/context/info have 5 ipv6 addresses....but there are none on guest if i type ifconfig 1154700861 M * Bertl Hollow: yep, sounds like a plan, make the redirects only on linux-vserver.org (so that you can access identical pages via wiki.linux-vserver.org and be done 1154700889 M * bonbons yang: ok, so chbind6 works as expected! Are the IPv6 addresses configured on some interface? They can only be visible to guests it they exist on the host 1154700924 M * Hollow Bertl: i.e. you want the new wiki at wiki.l-v.org and leave the old one on l-v.org? 1154700954 M * Bertl no, what I want is the following: 1154700993 M * Bertl - wiki pages and redirects happen on linux-vserver.org 1154701009 M * Bertl - no redirects (just pure wiki) happens on wiki.* 1154701043 M * Bertl does that make sense? 1154701062 M * Hollow yep, and the old wiki will go to old.* or whatever? 1154701074 M * Bertl yep, something like that 1154701076 M * yang bonbons: I have on "main" ipv6 addresses with 4 Ipv6 Ip's and i added the others to guest, using hte chbind6 1154701096 M * yang bonbons: on main the ipv6 addresses are on eth0 1154701119 M * bonbons so each address you added to the guests exists on the host? 1154701131 M * Hollow Bertl: ok, sounds good :) 1154701137 M * yang bonbons: no i added different addresses 1154701239 M * yang bonbons: http://paste.debian.net/10252 1154701321 M * bonbons from your output the guest should see this address: 2a01:b0:1114::25:3/64 1154701439 M * yang bonbons: and here ifconfig from the guest - http://paste.debian.net/10254 1154701589 M * bonbons hmm, could you check with ip (from iproute2)? 1154701602 Q * derjohn Ping timeout: 480 seconds 1154701612 Q * lilalinux Ping timeout: 480 seconds 1154701646 M * Hollow Bertl: regarding locked pages, at least some of the front pages should be locked imo 1154701811 Q * x03 Server closed connection 1154701814 M * yang bonbons: now i got this error ...Vserver 'vserver2' is running at context '49154' 1154701816 M * yang centaurus:/proc/virtnet# ls /proc/virtnet 1154701817 M * yang 49158 info 1154701836 M * yang two different context ID's 1154701841 J * x03 ~x03@213.58.38.139 1154701873 M * bonbons yang, are you using dynamic context IDs? 1154701883 M * yang i don't know? 1154701907 M * yang it happened when i restarted it 1154701961 M * bonbons do you have this file, and what is it's content: /etc/vservers//context ? 1154702017 M * yang centaurus:/etc/vservers/vserver2# ls 1154702017 M * yang apps fstab interfaces name run uts vdir 1154702099 J * mire_ ~mire@152-167-222-85.COOL.ADSL.VLine.Verat.NET 1154702121 J * lilalinux ~plasma@dslb-084-058-234-053.pools.arcor-ip.net 1154702132 J * derjohn ~derjohn@dslb-084-058-234-053.pools.arcor-ip.net 1154702181 M * bonbons so you are using dynamic contexts... stop the guest(s) and put a file names 'context' with unique context ID in it into each guest 1154702362 M * bonbons that way you will avoid possible issue where ID and name don't match or context ID changing after guest restart 1154702467 M * Bertl Hollow: hmm, how would that help? 1154702548 M * Hollow well, i think a f*cked up font page is not good PR ;) 1154702561 M * Hollow seems like i take an eye on PR now :P 1154702581 M * Bertl but a f* up non-frontpage is? 1154702638 M * yang centaurus:/etc/vservers# vserver vserver2 start 1154702639 M * yang vcontext: vc_create_context(): Invalid argument 1154702645 M * Bertl Hollow: I think we have to monitor page changes anyway 1154702672 M * Hollow sure we need to, but for the front page i think it would be a good idea, we can leave all other pages unlocked.. 1154702678 M * Hollow it's the first thing users see 1154702687 M * Bertl Hollow: and a locked fp only keeps folks from updating relevant info there, no? 1154702688 M * Hollow so it should be ok always 1154702698 M * Hollow well, which relevant info? 1154702710 M * Hollow (beside the version table and news) 1154702716 M * Bertl exactly that 1154702718 M * Hollow and these are templates and can be modified 1154702730 M * Bertl okay, then it's fine 1154702765 M * Bertl locking down the 'basic' structure is always okay 1154702790 M * Hollow sidebar is locked anyway, all other things can be left unlocked.. 1154702791 Q * click Server closed connection 1154702802 J * click click@ti511110a080-4165.bb.online.no 1154702850 M * bonbons yang: what context IDs did you choose? 1154702851 M * Roey Bertl, is ngnet going to be integrated into VServer in the next few months do you think? And once it does, how much easier will it be to run openvpn on it? 1154702859 M * bonbons best set them below 40000 1154703014 M * Bertl Roey: ngnet or virtualized networking will be part of Linux-VServer in the future ... 1154703160 J * coocoon ~coocoon@p54A05D3D.dip.t-dialin.net 1154703224 M * Bertl wb click! coocoon! 1154703242 M * coocoon high bertl 1154703363 M * Hollow Bertl: btw, any progres regarding accounting syscalls? 1154703405 M * Bertl IIRC, I uploaded the interface changes, but I think I will slightly revise it once again 1154703427 M * Bertl but the implementation is not a big deal, so should be done pretty soon 1154703433 M * Hollow ok, great 1154704619 Q * lilalinux Remote host closed the connection 1154704897 J * stefani ~stefani@tsipoor.banerian.org 1154705003 J * lilalinux ~plasma@dslb-084-058-234-053.pools.arcor-ip.net 1154705029 M * Bertl welcome stefani! lilalinux! 1154705041 M * stefani salut. 1154705198 M * Roey stefani, hi! 1154705548 J * sezuan matthias@agamemnon.ipv6.scheff32.de 1154705555 M * sezuan hi. 1154705669 M * sezuan I need more file handles in a vserver, I've increased the rlimits in the configuration and adjustes the /etc/security/limits.conf, but it seems that I can't increase the limit of the host system. 1154705814 M * sezuan oh, is that true that I must increase the hosts ulimits to have bigger ulimits in the host system? 1154705937 M * Bertl hmm, yes? 1154705982 M * Bertl but you probably meant the guest, so a yes there too, because ulimits are inherited .. so if your distro applies ulimits on the host, they will apply to the guests too, if not changed 1154706005 M * sezuan yes, your right. 1154706058 M * sezuan and with the rlimits I can limit the guests ulimits even further? or are rlimits totally different? 1154706091 Q * lilalinux Remote host closed the connection 1154706162 M * sezuan hmm.. rlimits seems to be global. 1154706169 J * lilalinux ~plasma@dslb-084-058-234-053.pools.arcor-ip.net 1154706225 Q * shedi Quit: Leaving 1154706243 M * yang bonbons: i will return later in the evening, thanks for helping me 1154706830 M * Bertl sezuan: the naming is a little confusing actually ... 1154706853 M * Bertl sezuan: the kernel knows about rlimits and we added per-guest-rlimits 1154706878 M * Bertl userspace calls the rlimits, ulimits (as do we) 1154706893 M * sezuan I start to understand :-) 1154706900 M * Bertl but we decided to call the per-guest-rlimits, just rlimits 1154706926 M * Bertl so, there are limits (in userspace called ulimits) which are per process basically 1154706949 M * Bertl and there are 'other' limits (in userspace called rlimits) which apply per guest 1154706965 M * Bertl to make that worse, there are hard and soft limits for both of them 1154706986 M * Bertl your filelimit probably comes from a hard limit on the host (see ulimit -aH 1154706987 M * sezuan I see. I set rlimit nproc to 900 and ulimit nproc to 1500, and I could not start more than 900 processes. 1154707008 M * Bertl precisely, because the rlimit is _binding_ per guest 1154707023 M * sezuan that's what I thought. 1154707088 M * sezuan All the stuff is a bit confusing. caps,ccaps,bcaps, ulimits(kernel 2.4),rlimit(2.6), legacy and new style, :-) 1154707187 M * sezuan btw, the wiki says: Context Flags (vs2.0) ' INFO_ULIMIT ulimit L apply ulimits to context' L=will disappear? 1154707192 M * Bertl yeah, it has grown :) 1154707222 M * Bertl the apply ulimit is probably gone already (at least for recent tools) 1154707272 M * sezuan ok. and one final question. When I want to limit the memory usage on a guest system, I think I should know what's the difference between the RSS and AS limits? 1154707369 M * Bertl yes, and probably for other memory related limits too 1154707389 M * Bertl basically AS means Address Space, while RSS means Resident Set Size 1154707508 M * sezuan hm.. what should I do if a want a machine with the equivalent of 128MB real memory an 512MB swap? Or is it not possible to translate that in rlimits? 1154707524 M * sezuan machine should be guest. 1154707587 Q * _Kara Quit: just gone 1154707943 M * Bertl it's not simple to translate that, because there is a lot of memory sharing going on 1154708164 M * Bertl to make it 'look' like those are the limits of the guest, assign 640M RSS hard limit and a 128M soft limit 1154708208 M * Bertl (i.e. that will show up as mem and swap respectively) 1154708284 M * sezuan thanks. 1154708660 M * Bertl you're welcome! 1154708987 Q * sladen Ping timeout: 480 seconds 1154709139 J * sladen paul@starsky.19inch.net 1154709912 M * sezuan http://linux-vserver.org/Resource+Limits, Am I right that the limits with "U-" can't be used with 2.6 Kernels? 1154710010 M * Bertl they can be used as ulimits (that's what the U says) but they are not supported as per guest rlimits (the '-') 1154710034 M * Bertl but most of them do not make much sense as rlimit anyway 1154710062 M * sezuan ok. /etv/vservers//ulimits won't be used for 2.6 kernels? 1154710066 M * Bertl i.e. you probably don't want your guest to run for 10 seconds and be killed off after that (for example) 1154710099 M * Bertl sezuan: it should be used with recent tools, but it will not override the host limits 1154710400 J * EKC ~EKC@netblock-66-245-252-180.dslextreme.com 1154710434 M * Bertl welcome EKC! 1154710460 M * EKC howdy 1154710613 Q * bubulak Server closed connection 1154710625 J * bubulak ~bubulak@whisky.pendo.sk 1154710724 Q * Smutje Server closed connection 1154710726 J * Smutje ~Smutje@xdsl-87-78-98-134.netcologne.de 1154710844 J * insomniac ~insomniac@slackware.it 1154711852 J * kevinp ~kevinp@ny.webpipe.net 1154711931 M * kevinp so was there a workaround in the new code that allows bind to work inside a vserver without changes caps? 1154711966 M * kevinp I had one vserver on one box start a default install of bind with no problem with no changes to caps 1154711968 M * Bertl welcome kevinp! and yes! 1154712001 M * kevinp but then I upgraded another vserver to the latest kernel patches and I still got an error inside the vserver 1154712032 M * Bertl probably a stable kernel patch? 1154712034 M * kevinp Here's the info on the two servers: http://deadbeefbabe.org/paste/1413 1154712053 M * kevinp Yeah, I think I grabbed the latest stable 1154712090 M * Bertl yep, the paste shows, this version doesn't have this new feature 1154712111 M * Bertl (while the 2.1.1 branch has) 1154712152 M * kevinp brb 1154712513 M * kevinp sorry, I didn't catch that, for some reason I thought I was getting 2.1.2 and not 2.0.2 1154712532 M * kevinp and that you guys had been really busy getting to 2.1.2 already :) 1154712576 M * Bertl well, no problem, we'll get to 2.1.2 soon :) 1154712588 M * kevinp you're the best, thanks! 1154712924 P * kevinp Compiling... 1154713079 Q * phedny Server closed connection 1154713089 J * phedny ~mark@volcano.p-bierman.nl 1154714024 J * Blissex ~Blissex@82-69-39-138.dsl.in-addr.zen.co.uk 1154714803 M * Bertl wb phedny! Blissex! 1154714824 A * Blissex waves hi! 1154716440 M * EKC instead of using 'vserver stop', is there a way to stop a vserver using only the context id? 1154716496 Q * Savvy Remote host closed the connection 1154716522 Q * jake- Server closed connection 1154716530 M * Bertl EKC, yes, the tools can figure the name from the xid too, but maybe your actual question is something like: "I removed my guest and now I need to stop processes in context xy?" 1154716537 J * jake- psybnc@murlocs.org 1154716543 M * Bertl welcome jake-! 1154716573 M * EKC yes, that should have been my question :) any advice? vkill? 1154716588 M * Bertl yep, vkill works nicely here 1154716603 M * EKC but, also, what would the command be to map an xid to a vserver name? 1154716654 M * Bertl vserver-info 1154716685 M * EKC ah. yes. thx 1154717008 M * Bertl okay, off for now ... back later this evening! 1154717012 N * Bertl Bertl_oO 1154717611 J * shedi ~siggi@inferno.lhi.is 1154718573 Q * meebey Server closed connection 1154718597 J * meebey meebey@booster.qnetp.net 1154718773 M * cehteh http://www.heise.de/newsticker/meldung/76389/from/rss09 .. where is our PR department? 1154718996 Q * EKC Remote host closed the connection 1154720548 J * Hurga nobody@p508A9429.dip0.t-ipconnect.de 1154720622 Q * bogus Server closed connection 1154720622 J * bogus_ ~bogusano@fengor.net 1154720644 M * Hurga Hi. Quick question: what do I need for per-context quota with the current version of vserver? It's been some time since I used it... 1154720964 M * Hurga hello...? 1154721465 A * Hurga sighs. 1154721787 Q * Greek0 Quit: Lost terminal 1154721822 J * Greek0 ~greek0@85.255.145.201 1154722106 A * Hurga tries to wait patiently... 1154722907 Q * insomniac Quit: Lost terminal 1154723043 J * dna_ ~naucki@211-222-dsl.kielnet.net 1154723173 Q * pisc1 Quit: Download Gaim: http://gaim.sourceforge.net/ 1154723219 M * bonbons Hurga: Take a look here http://linux-vserver.org/Quota+and+Disk+Limits 1154723258 M * bonbons I can't help you much more as I never used disk quota (neither with nor without vserver) 1154723312 Q * dna Ping timeout: 480 seconds 1154723336 M * Hurga I get the impression that per context disk quota is not yet supported in 2.6 :( 1154723581 M * bonbons Hurga, it's certainly better supported as on 2.4, but for details you would probably have to wait for someone using it... 1154723657 M * bonbons other possible source of information: http://linux-vserver.org/Disk+Limits 1154723974 Q * Greek0 Quit: leaving 1154723987 J * Greek0 ~greek0@85.255.145.201 1154724457 J * insomniac ~insomniac@slackware.it 1154725680 Q * dna_ Quit: Verlassend 1154727085 Q * bonbons Quit: Leaving 1154727617 J * enet ~jpduyx@adsl-228-22.dsl.uva.nl 1154727656 M * Hurga Hi 1154727661 M * enet hi 1154727698 M * Hurga do you happen to use vserver with quota? :) 1154727726 M * enet mmmm, sorry, using vserver but notyet with quota 1154727759 A * Hurga currently trying to understand the startup sequence 1154728227 P * enet 1154728289 M * Skram hey 1154728395 M * Hurga Hi 1154728699 J * schimmi ~sts@brs9-d9baf098.pool.mediaWays.net 1154729057 M * Hurga How do I set capabilities with the new configuration system? 1154729076 M * Skram AFAIK... /etc/vservers/VSERVER_NAME/Xcapabilities 1154729084 M * Skram if you check out the wiki.. you will see, I believe 1154729230 M * Hurga well, I've been trying for that last haf hour or so to find info about it... 1154729305 M * Skram you restart the VServer after making changes, correct/ 1154729340 M * Hurga So far, I don't even know where to make the changes :) 1154729584 M * Hurga I mean, the FAQ tells me "There are a number of capabilities and flags that you can modify listed here" but not where in the config I would do that. 1154729627 M * Hurga no hits for "Xcapabilities" BTW. 1154729710 M * coocoon Hurga: 1154729725 M * coocoon Hurga: which distro are u using and what will u do exactly 1154729740 M * coocoon Hurga: i mean where is the problem 1154729764 M * Skram coocoon: well, for example some go in acapabilities, bcapabilities 1154729767 M * Skram sorry for the confusion 1154729835 M * Hurga Skram: I've been using the old config method so far, i don't even know what these [a-z]capabilities are. 1154729909 M * Hurga coocoon: I have an ancient vserver running bind. I want to migrate it. I know this is an FAQ and that named needs CAP_SYS_RESOURCE, but I can't find how to set it. 1154730009 M * coocoon Hurga: u have configured the kernel for using vroot devices and they are there 1154730009 M * Hurga in the old config I needed to do S_CAPS="CAP_SYS_RESOURCE" 1154730034 M * Hurga coocoon: I'm using the Ubuntu pre-pakaged vserver kernel 1154730054 M * coocoon Hurga: i do not know it exactly but i have never used the config options with kernel 2.6.x i meant this is only needed for 2.4.x kernel 1154730124 M * coocoon Hurga: maybe this will help http://www.5dollarwhitebox.org/wiki/index.php/Howtos_Linux-Vserver_With_LVM_And_Quotas 1154730138 M * coocoon Hurga: it is for debian ;-) 1154730171 Q * Blissex Remote host closed the connection 1154730246 M * coocoon Hurga: but also there is /etc/vservers.conf where u can set something, but as i said have never used it before 1154730336 M * Hurga ok, http://www.nongnu.org/util-vserver/doc/conf/configuration.html is what I was looking for. 1154730361 M * coocoon Hurga: yes right 1154730362 M * Hurga unfortunately, some sadist made it almost unreadable... 1154730405 M * coocoon Hurga: hehe 1154730507 J * yarihm ~yarihm@84-74-17-70.dclient.hispeed.ch 1154731447 Q * Nam Ping timeout: 480 seconds 1154732469 P * stefani I'm Parting (the water) 1154732659 Q * mire_ Quit: Leaving 1154734607 Q * coocoon Ping timeout: 480 seconds 1154735183 J * coocoon ~coocoon@p54A07F32.dip.t-dialin.net