1187049607 J * buzztracker ~buzztrack@pelikan.garga.net 1187049655 M * cehteh Bertl: do you know the block io layer well? .. i found http://www.acis.ufl.edu/~ming/dmcache/ .. it would be interesting to turn that suitable for caching read accessed blocks on a laotop in some kind of flash memory 1187049691 M * daniel_hozac RAM's not good enough for you? :) 1187049735 M * cehteh daniel_hozac: nope .. i maxed my laptop with 1GB and the disk often needs to spin up for reads 1187049743 M * cehteh writes are handled fine by laptop mode 1187049805 M * cehteh with 4-8GB CF cards which have 20-40MB sec access times it would be reall nice to use such as cache .. 1187049831 M * fatgoose cf card for caching? 1187049838 M * cehteh caching on flash just needs very different caching policies than that what the dmcache currently does 1187049872 M * cehteh fatgoose: just any kind of flash device, usb stick, sd card, cf .. whatever you like 1187049894 M * cehteh for me CF would be the best since my pc-card slot is unused 1187049902 M * fatgoose but cf have limited read/write count, right? 1187049905 M * cehteh yes 1187049928 M * cehteh hence diffrent policies and hence i bringing up the question 1187049942 M * fatgoose your harddrive is not faster? 1187049942 M * cehteh you only want to cache the most frequent readonly accesses 1187049973 M * fatgoose you better buy more ram than cache on a CF 1187049975 M * cehteh laptop hd? no .. and it has latency plus needs some time to spin up when in sowersafe 1187049983 M * fatgoose 8GB is cheap 1187049990 M * fatgoose oh your laptop 1187049992 M * cehteh my laptop can only handle 1GB 1187050060 M * cehteh hey i think i know what i am talking about :P what i dont know is if the kernel could see if a blockio request is readonly or not (but i didnt investigated it yet) 1187050120 M * cehteh the caching algo should prevent writes to CF at all costs 1187050165 M * cehteh means most frequently readonly accesses (needs some way to figure them out) and only on page disposal, not at load 1187050203 J * esa ~esa@ip-87-238-2-45.adsl.cheapnet.it 1187050214 M * daniel_hozac the most frequent ones would be kept in RAM :) 1187050265 M * cehteh nope .. when saying 'read' i meant read, not accessed (from pagecace) 1187050359 M * cehteh such media cant be mmaped in place or? 1187050396 M * cehteh (sstupid question, of course not :P) 1187051080 Q * esa Ping timeout: 480 seconds 1187053193 Q * buzztracker Quit: Leaving 1187053202 J * buzztracker ~buzztrack@pelikan.garga.net 1187053301 M * Bertl okay, translocating .. back later .. 1187053308 N * Bertl Bertl_oO 1187056792 Q * buzztracker Quit: Leaving 1187056806 J * buzztracker ~buzztrack@pelikan.garga.net 1187057007 Q * fatgoose Quit: fatgoose 1187057600 J * Freax ~linus@bl7-132-35.dsl.telepac.pt 1187058033 Q * Linus Ping timeout: 480 seconds 1187060396 Q * buzztracker Quit: Leaving 1187060403 J * buzztracker ~buzztrack@pelikan.garga.net 1187061553 N * Bertl_oO Bertl 1187061626 F * ChanServ +o Bertl 1187061647 F * Bertl +b *!*buzztrack@*.garga.net 1187061674 M * Bertl buzztracker: please contact me if you want to get un-banned ... 1187061692 F * Bertl -o Bertl 1187062368 Q * Baby Remote host closed the connection 1187063993 Q * buzztracker Quit: Leaving 1187064439 Q * Aiken Ping timeout: 480 seconds 1187066570 M * Bertl okay, off to bed for today ... have a good one everyone! 1187066583 N * Bertl Bertl_zZ 1187067444 N * AstralSt AStorm 1187069634 Q * matti Ping timeout: 480 seconds 1187071635 N * DoberMann_ DoberMann 1187073474 J * Risjh rsjds@59.93.134.171 1187073656 N * DoberMann DoberMann[PullA] 1187074250 J * esa ~esa@ip-87-238-2-45.adsl.cheapnet.it 1187075073 J * dna ~dna@216-248-dsl.kielnet.net 1187075143 Q * FBW Quit: leaving 1187076741 Q * dilinger Remote host closed the connection 1187077032 J * sharkjaw ~gab@158.36.44.106 1187077220 Q * rob-84x^ Ping timeout: 480 seconds 1187077670 Q * coderanger_ Quit: coderanger_ 1187077861 J * marcfiu ~marcfiu@c-68-39-177-97.hsd1.nj.comcast.net 1187077873 Q * marcfiu 1187078763 Q * aj_ Ping timeout: 480 seconds 1187079183 J * dilinger ~dilinger@mail.queued.net 1187080350 N * fb_ fb 1187080966 J * Baby ~miry@195.37.62.208 1187081924 J * rob-84x^ rob@submarine.ath.cx 1187083474 J * coderanger_ ~coderange@1cc-dhcp-120.media.mit.edu 1187086634 J * meandtheshel1 ~markus@85.127.103.154 1187086716 J * Aiken ~james@ppp121-45-255-55.lns2.bne4.internode.on.net 1187087074 Q * Freax Remote host closed the connection 1187088185 M * transacid ooh when did 2.6.22.2-vs2.2.0.3 come out? 1187088758 Q * Aiken Remote host closed the connection 1187088971 J * Aiken ~james@ppp121-45-255-55.lns2.bne4.internode.on.net 1187090487 Q * coderanger_ Quit: coderanger_ 1187091528 Q * transacid Remote host closed the connection 1187091889 J * transacid ~transacid@transacid.de 1187092645 Q * ensc Ping timeout: 480 seconds 1187093648 J * Aiken_ ~james@ppp121-45-255-55.lns2.bne4.internode.on.net 1187093664 Q * Aiken Read error: Connection reset by peer 1187094272 J * Tesla2k ~jascha@p549C610A.dip.t-dialin.net 1187094399 N * sladen_ sladen 1187095224 M * eyck hmm 1187095247 M * eyck I got a BUG in kernel, and the google only suggested something on #vserver being talked about 1187095250 M * eyck it looks like this: 1187095252 M * eyck EIP: [] dev_in_nx_info+0xe/0x6c SS:ESP 0068:cf889e9c 1187095280 M * eyck BUG: unable to handle kernel NULL pointer dereference at virtual address 000000ac 1187095299 M * eyck is it possible that it's related to vserver? 1187095988 M * daniel_hozac well, yeah. that's a vserver function :) 1187096003 M * daniel_hozac whole bug on paste.linux-vserver.org please. 1187096049 Q * sladen Ping timeout: 480 seconds 1187096064 J * sladen paul@starsky.19inch.net 1187096460 M * transacid there has been a lot of spam on the paste.linux-vserver.org 1187096774 M * eyck hmm, this is it: http://paste.linux-vserver.org/pastebin.php?dl=5597 ? 1187097339 P * friendly12345 1187097346 M * eyck it's on 2.6.18 with vs2.0.3 1187097360 M * daniel_hozac looks like an old bug that is long since fixed. 1187097494 M * daniel_hozac even recent 2.0.3-rc's have that fix. 1187097818 J * gerrit ~gerrit@dslb-084-060-232-164.pools.arcor-ip.net 1187097938 J * matti matti@acrux.romke.net 1187098310 N * Bertl_zZ Bertl 1187098318 M * Bertl morning folks! 1187098324 M * daniel_hozac morning Bertl! 1187098741 M * Bertl hey daniel_hozac how's going? 1187098874 M * daniel_hozac good, a bit tired, thanks. and you? 1187098915 M * bXi morning 1187098956 M * Bertl I slept quite well, I think I'm fine 1187099049 M * daniel_hozac nice 1187099274 M * bXi i guess its not possible to run a solaris 10 guest 1187099281 A * bXi knows nothing about solaris 1187099320 M * daniel_hozac if your kernel supports the binaries, it should be fine, i think. 1187099328 M * daniel_hozac (for varying degrees of fine) 1187099351 M * bXi there was this action so one could receive free solaris 10 licenses 1187099360 M * bXi so i obviously went allong with that 1187099365 Q * Tesla2k Quit: Verlassend 1187099403 M * daniel_hozac hmm, i thought solaris was free4all these days. 1187099412 M * bXi hmmm 1187099422 M * bXi i think i'm getting free cd's and everything with it 1187099430 M * daniel_hozac ah. 1187099442 M * bXi should be here around next week 1187099517 M * bXi quite curious on how this works and feels 1187099634 M * bXi hmm the free dvd kit offer has expired 1187099725 M * Bertl well, IIRC, it feels slow, unless you got a fast machine :) 1187099748 M * bXi depends 1187099760 M * bXi is an amd x2 64 4000+ slow (relatively obviously) 1187099851 M * Bertl well, solaris on intel is slower by default but not as slow as SCO :) 1187099865 Q * gerrit Read error: Connection reset by peer 1187100007 M * Bertl okay, off to the shower .. back shortly ... 1187100025 N * Bertl Bertl_oO 1187100506 J * fatgoose ~samuel@76-10-147-136.dsl.teksavvy.com 1187101133 J * rgl ~rgl@84.90.10.107 1187101138 M * rgl hello 1187101143 M * daniel_hozac hi 1187101173 M * rgl we can use IPv6 in 2.2.x? or do I need 2.3.x? 1187101243 M * daniel_hozac it's a separate patch for 2.2. 1187101261 M * daniel_hozac 2.3 has it included. 1187101439 M * rgl thx daniel_hozac . 1187101466 M * rgl http://people.linux-vserver.org/~dhozac/p/k/patch-2.6.20.15-vs2.2.0.3-ipv6.diff is the latest one? there is none for 2.6.22.2? 1187101503 M * daniel_hozac no, i haven't made one yet. 1187101506 J * bere ~bere@nat65.homenet.glazov.net 1187101516 M * rgl I can try 2.3.x then :) 1187101526 M * daniel_hozac sounds good :) 1187101531 M * bere hi ppl 1187101543 N * Bertl_oO Bertl 1187101546 M * daniel_hozac (as long as it's not production, ) 1187101550 M * daniel_hozac hello bere 1187101552 M * Bertl back now! hey bere! 1187101569 M * bere i havbe a problem with pam_vserver 1187101590 M * bere problem with devices and /proc 1187101592 M * daniel_hozac (hmm, i thought that was just an idea someone had...) 1187101606 M * rgl do I have to use a specific util-vserver? or 0.30.213 will do? 1187101611 M * daniel_hozac anyway, what kind of problems? 1187101622 M * daniel_hozac rgl: http://people.linux-vserver.org/~dhozac/t/uv-testing/util-vserver-0.30.214-pre2585.tar.bz2 1187101650 M * bere i need at analog of pam_chroot ( 1187101670 M * rgl daniel_hozac, thx. later today I'll try it :) 1187101770 M * rgl daniel_hozac, will that util-vserver work with 2.2.x? 1187101786 M * daniel_hozac it should work with pretty much any kernel. 1187101815 M * bere every user what connect to server by ssh must be palced in vserver 1187101828 M * bere placed* 1187101853 M * bere how do this? 1187101860 M * rgl bere, place the sshd daemon inside the guest? 1187101876 M * bere i have a 3 vserver on one ipaddress 1187101886 M * bere i acnnot change the default ssh port 1187101900 M * daniel_hozac that's kinda icky, but okay. 1187101906 M * bere and i must can connect to host server 1187101921 M * daniel_hozac what problems are you having with pam_vserver (and where's the source?)? 1187101959 M * bere with pam_server a have problem with broken link from /dev/fd 1187101977 M * bere and i not have mount'ed proc 1187101978 M * bere but 1187101990 M * bere even undo root i cannot mount this! 1187101999 M * daniel_hozac as is expected... 1187102006 M * bere why? 1187102024 M * daniel_hozac you need capabilities to be able mount anything. 1187102040 M * daniel_hozac the _real_ question is, why isn't it already mounted? 1187102064 M * daniel_hozac sounds to me like pam_vserver isn't really doing what it should. 1187102110 M * bere what a cap i must enabled? 1187102124 M * Bertl for normal operation, you do not need any caps 1187102139 M * bere for mount... 1187102147 M * Bertl adding capabilities lowers your security and possibly compromises your guest 1187102147 M * daniel_hozac make pam_vserver do the right thing instead. 1187102168 M * Bertl if you still want to do that, here is the information 1187102197 M * Bertl http://linux-vserver.org/Capabilities_and_Flags 1187102204 M * bere thx 1187102442 M * daniel_hozac note that this is probably just one symptom of a much larger issue. 1187102461 J * gerrit ~gerrit@bi-02pt1.bluebird.ibm.com 1187102475 M * Bertl I would opt for a missing vprocunhide :) 1187102504 M * daniel_hozac i'm actually more inclined to blame pam_vserver, as even /proc/self/fd didn't exist. 1187102524 Q * Risjh Ping timeout: 480 seconds 1187102531 M * Bertl could be ... 1187102574 Q * sharkjaw Remote host closed the connection 1187102772 Q * Aiken_ Remote host closed the connection 1187102783 J * Aiken_ ~james@ppp121-45-255-55.lns2.bne4.internode.on.net 1187103397 J * Piet ~piet@tor.noreply.org 1187103608 M * Bertl okay, off to grab a few things .. should be back in the afternoon 1187103638 N * Bertl Bertl_oO 1187104010 M * transacid Bertl_oO: now i know what that buzztracker is you banned: http://www.buzztracker.org/about/ 1187104020 Q * Aiken_ Quit: Leaving 1187105048 M * bere what this : "Asking all remaining processes to terminate...done. 1187105048 M * bere Killing all remaining processes...done. 1187105048 M * bere cat: /proc/cmdline: No such file or directory" 1187105399 M * daniel_hozac your guest's initscripts are accessing files which are hidden from guests. 1187105471 M * bere how it cure? 1187105479 M * bere debian 4.0 ) 1187105487 M * bere util-vbserver from packet 1187105538 M * daniel_hozac it's just a cosmetical thing. just ignore it. 1187105794 M * bere with this pam_bserver i have a probelm (i say this) that even undo root i cannot anything mount - mount: permission denied; if i set SYS_ADMIN - it help me ? 1187105830 M * bere this aim decide my problem? 1187105893 N * Bertl_oO Bertl 1187105974 M * Bertl bere: as we alreadz pointed out, if you give SYS_ADMIN, you basically allow the guest to mess with your host system .. if that is what you want .. go ahead 1187105982 M * Bertl *already 1187106002 Q * Baby cation.oftc.net charon.oftc.net 1187106002 Q * dna cation.oftc.net charon.oftc.net 1187106002 Q * fb cation.oftc.net charon.oftc.net 1187106002 Q * DoberMann[PullA] cation.oftc.net charon.oftc.net 1187106002 Q * derjohn cation.oftc.net charon.oftc.net 1187106002 Q * nebuchadnezzar cation.oftc.net charon.oftc.net 1187106002 Q * vasko cation.oftc.net charon.oftc.net 1187106002 Q * DuckMaster cation.oftc.net charon.oftc.net 1187106002 Q * harry cation.oftc.net charon.oftc.net 1187106002 Q * Loki|muh cation.oftc.net charon.oftc.net 1187106002 Q * Guy- cation.oftc.net charon.oftc.net 1187106002 Q * Vudu cation.oftc.net charon.oftc.net 1187106002 Q * Greek0 cation.oftc.net charon.oftc.net 1187106002 Q * FloodServ cation.oftc.net charon.oftc.net 1187106002 Q * mugwump cation.oftc.net charon.oftc.net 1187106002 Q * hallyn cation.oftc.net charon.oftc.net 1187106002 Q * tam cation.oftc.net charon.oftc.net 1187106002 Q * djbclark cation.oftc.net charon.oftc.net 1187106002 Q * Supaplex cation.oftc.net charon.oftc.net 1187106002 Q * fatgoose cation.oftc.net charon.oftc.net 1187106002 Q * dilinger cation.oftc.net charon.oftc.net 1187106002 Q * mountie cation.oftc.net charon.oftc.net 1187106002 Q * ntrs__ cation.oftc.net charon.oftc.net 1187106002 Q * Hollow cation.oftc.net charon.oftc.net 1187106002 Q * dos000 cation.oftc.net charon.oftc.net 1187106002 Q * brcc_ cation.oftc.net charon.oftc.net 1187106002 Q * coderanger cation.oftc.net charon.oftc.net 1187106002 Q * gdm cation.oftc.net charon.oftc.net 1187106002 Q * neuralis cation.oftc.net charon.oftc.net 1187106002 Q * balbir__ cation.oftc.net charon.oftc.net 1187106002 Q * mattzerah cation.oftc.net charon.oftc.net 1187106002 Q * Johnnie cation.oftc.net charon.oftc.net 1187106002 Q * micah cation.oftc.net charon.oftc.net 1187106002 Q * quasisane cation.oftc.net charon.oftc.net 1187106002 Q * hardwire cation.oftc.net charon.oftc.net 1187106002 Q * bored2sleep cation.oftc.net charon.oftc.net 1187106032 J * fatgoose ~samuel@76-10-147-136.dsl.teksavvy.com 1187106032 J * Baby ~miry@195.37.62.208 1187106032 J * dilinger ~dilinger@mail.queued.net 1187106032 J * dna ~dna@216-248-dsl.kielnet.net 1187106032 J * fb fback@red.fback.net 1187106032 J * DoberMann[PullA] ~james@AToulouse-156-1-21-253.w86-196.abo.wanadoo.fr 1187106032 J * mountie ~mountie@trb229.travel-net.com 1187106032 J * balbir__ ~balbir@122.167.93.202 1187106032 J * Johnnie ~jdlewis@c-67-163-142-234.hsd1.ct.comcast.net 1187106032 J * ntrs__ ~ntrs@68-188-55-120.dhcp.stls.mo.charter.com 1187106032 J * derjohn ~derjohn@80.69.41.3 1187106032 J * nebuchadnezzar ~nebu@zion.asgardr.info 1187106032 J * djbclark dclark@opensysadmin.com 1187106032 J * vasko ~vasko@unreal.rainside.sk 1187106032 J * Hollow ~hollow@78.47.240.174 1187106032 J * DuckMaster ~Duck@tox.dyndns.org 1187106032 J * dos000 ~dos000@CPE000f66912f92-CM0018c0c6147e.cpe.net.cable.rogers.com 1187106032 J * brcc_ bruce@72.20.27.65 1187106032 J * coderanger ~coderange@c-65-96-210-168.hsd1.ma.comcast.net 1187106032 J * mugwump ~samv@watts.utsl.gen.nz 1187106032 J * hallyn ~xa@adsl-75-0-151-14.dsl.chcgil.sbcglobal.net 1187106032 J * harry ~harry@d54C2508C.access.telenet.be 1187106032 J * micah ~micah@micah.riseup.net 1187106032 J * quasisane ~sanep@c-75-67-252-214.hsd1.nh.comcast.net 1187106032 J * Supaplex supaplex@166-70-62-199.ip.xmission.com 1187106032 J * FloodServ services@services.oftc.net 1187106032 J * Vudu ~vudumen@perverz.hu 1187106032 J * Guy- VpFTQTAnHs@chardonnay.math.bme.hu 1187106032 J * Greek0 ~greek0@85.255.145.201 1187106032 J * bored2sleep ~bored2sle@66.111.53.150 1187106032 J * Loki|muh loki@satanix.de 1187106032 J * hardwire ~bip@216.152.176.9 1187106032 J * tam ~tam@gw.nettam.com 1187106032 J * neuralis ~krstic@solarsail.hcs.harvard.edu 1187106032 J * gdm ~gdm@www.iteration.org 1187106032 J * mattzerah ~matt@121.50.222.55 1187106079 M * bere yes i understand what this not good idea...but what me do ??? 1187106120 M * daniel_hozac fix pam_vserver? or complain to whoever did write it? 1187106160 M * Bertl probably there is some incompatibility between this software and the actual kernel/tools you are using 1187106176 M * Bertl something like missing namespaces or so come to my mind 1187106192 M * daniel_hozac yeah, that's what i've been thinking too. 1187106241 M * Bertl bere: what kernel/tools and where is that pam_vserver from? 1187106342 M * bere util-vserver - 0.30.221-6 ; kernel - 2.6.22.1; pam_vserver - 0.0.1 (http://sourceforge.net/project/showfiles.php?group_id=179379) 1187106362 M * bere pam_vserver not update 1187106372 M * Bertl whoa 0.30.221, are you from the future? 1187106402 M * bere yes :)) 1187106420 M * bere this from debian 4 1187106453 M * daniel_hozac Debian who are usually behind... 1187106485 M * Bertl well, they might still be far behind if he is from the future :) 1187106492 M * bere debian testing ?) 1187106503 M * bere behind? 1187106509 M * bere hz hz ) 1187106509 M * Bertl daniel_hozac: do we have any pending changes to shiny? 1187106513 M * daniel_hozac haha, yeah, that's true. 1187106539 M * daniel_hozac Bertl: other than the, uh, whatever arch it was, being tested, none that i can think of. 1187106540 M * Bertl (except for the arm is now tested part) 1187106540 M * bXi 221? 1187106553 M * bere yes 1187106561 M * bXi i think you mean 212 1187106569 M * Bertl daniel_hozac: okay, let's remove that warning then and be fine, yeah? 1187106577 M * daniel_hozac yep, sounds good to me. 1187106599 M * bXi are there any new patches yet? 1187106606 M * bere uupccc 1187106608 M * bere sorry( 1187106612 M * bere i stupid ) 1187106616 M * bere 211-6 1187106618 M * bere )) 1187106638 M * bXi venice .openttd # vserver --version 1187106638 M * bXi vserver 0.30.214-pre2585 -- manages the state of vservers 1187106641 M * daniel_hozac now that's more Debian-like, 2 releases behind :) 1187106654 M * bXi venice .openttd # uname -a 1187106654 M * bXi Linux venice 2.6.22.2-vs2.3.0.17 #1 Sun Aug 12 13:08:45 CEST 2007 x86_64 AMD Athlon(tm) 64 Processor 3000+ AuthenticAMD GNU/Linux 1187106680 M * bXi daniel_hozac Bertl no errors so far 1187106682 M * bere daniel - any idea for place user in vserver?) 1187106685 M * daniel_hozac bXi: just my patches for multiple binders to the same port. 1187106704 M * daniel_hozac bXi: (in different guests) 1187106729 M * bXi as in multiple guests using the same ip? 1187106741 M * daniel_hozac no, as in multiple guests running sshd on IPv6. 1187106785 M * bXi ah 1187106790 M * bXi i'll be needing those me thinks 1187106816 N * virtuoso_ virtuoso 1187106965 M * bere in vserver i can disabled cap form this vserver ? 1187106971 M * bere for* 1187107007 M * daniel_hozac what? 1187107046 M * Bertl bere: there is a (quite old) trampoline script out there 1187107058 M * Bertl bere: maybe you can adjust that to your purposes 1187107083 M * Bertl http://www.paul.sladen.org/vserver/archives/200503/0068.html 1187107149 J * Risjh rsjds@59.93.135.222 1187107180 M * Bertl wb Risjh 1187107250 M * bere for vsserver1 i set sys_admin; i enter in vserver1 and _can_ i disable sys_admin for itself (vserver1) ? i think thatn no ) 1187107304 M * daniel_hozac you mean, can root inside the guest drop SYS_ADMIN? 1187107308 M * Bertl you can drop that capability, yes, but running any suid app will give it back 1187107442 M * bere daniel - yes 1187107502 M * bere bertl - so working application will have cap while app do not kill ? 1187107511 M * bere sorry for my bad english ) 1187107555 M * Risjh ty Bertl :) 1187107582 M * Bertl bere: sorry, please rephrase 1187107728 M * bere if i drop sys_admin and run suid app - it will have sys_admin ? 1187107773 J * arcil ~arcil@p5B077ED4.dip.t-dialin.net 1187107787 M * Bertl bere: exactly! 1187107792 M * Bertl wb arcil! 1187107855 M * bere so... senselessly drop cap in guest system? 1187107895 M * Bertl if you allow whatever is executing to start a suid app, then it doesn't make much sense, yes 1187107939 M * Bertl but check out the trampiline stuff, it might be old and outdated, but it should be no big deal to make it work with recent systems 1187108014 M * bere yes ) 1187108031 M * arcil ty Bertl 1187108662 M * eyck where can I find vs2.0.3 for 2.6.18? 1187108691 M * eyck or, how stable is 2.2.x? 1187108703 M * daniel_hozac 2.2 is stable. 1187108723 M * nanonyme heh 1187108750 M * nanonyme eyck: i've been told the numbering is the same as in linux kernel: even is stable and odd is experimental 1187108786 M * eyck daniel_hozac: what can I expect to brake when moving from vs2.0 to vs2.2? 1187108799 M * daniel_hozac nothing, assuming you have recent utils. 1187108804 M * eyck ooh. 1187108811 M * Bertl yeah, well, we call the odd ones development, and the 4 number versions experimental 1187108817 M * nanonyme heh 1187108821 M * eyck in other words, everything :) 1187108826 M * daniel_hozac 2.2.0.3 is experimental? :) 1187108828 M * eyck which utils do I need for 2.2? 1187108848 M * daniel_hozac 0.30.213+ recommended, but 0.30.210+ is required. 1187108863 M * eyck ok, I'm already at 213. thanx. 1187108863 M * nanonyme Bertl: the difference between development and experimental isn't really that clear in my mind so excuse me for sometimes using perhaps an unorthodox term :) 1187108883 M * Bertl daniel_hozac: I meant the 4 number, second odd versions :) 1187108923 M * daniel_hozac nanonyme: experimental means "this compiled for me", development means "this seems to work alright" 1187108955 M * daniel_hozac (greatly simplified) 1187108955 M * eyck was there any newer 2.0 then 2.0.3-rc3? 1187108965 M * daniel_hozac eyck: nope, that's the most recent one. 1187108966 M * Bertl actually experimental here means: it compiled for me, booted in qemu and didn't blow up with testme.sh :) 1187109002 M * Bertl which is more than you can expect from mainline :) 1187109005 M * nanonyme hehe 1187109031 M * eyck thanks. 1187109111 A * nanonyme grumbles at 2.4 linux kernel not handling ip6tables properly 1187109156 M * bere on my 2.6.22.1 with framebuffer and videocard gefroce 6600gt with vserver system hung; if id isabled fb in kernel system work god....it normal?) 1187109170 M * bere if i* 1187109180 M * bere if i disabled* 1187109181 M * bere )) 1187109190 M * Bertl did you try with a vanilla kernel to verify that it is Linux-VServer related? 1187109197 A * bXi builds yet another guest 1187109200 M * bere vanilla kernel 1187109222 M * bere without vserver fb work good 1187109238 M * bere on 2.6.18 no problem 1187109241 M * Bertl same config? then it is a bug 1187109251 M * daniel_hozac well, that's an apples and oranges comparison. 1187109253 M * bXi bere: try 2.6.22.1 without any patches 1187109265 M * daniel_hozac actually, try 2.6.22.2 without any patches. 1187109275 M * bere vanilla kernel 2.6.20 + patch .2.6.21 + patch 2.6.22 1187109275 M * Bertl bere: and copy the .config from the vserver kernel 1187109292 M * Bertl (the one which failed) 1187109306 M * bere these patches and kernel from kernel.org... 1187109328 M * Bertl yep, should be fine, you want to go for 2.6.22.2 though (one more patch) 1187109355 M * bXi isnt that a bad idea in this case? 1187109365 M * bXi since his error occurs on 2.6.22.1 ? 1187109381 M * Bertl np with testing with 2.6.22.1 either 1187109437 M * bXi Bertl: act of lazyness 1187109439 M * bXi venice ~ # nvserver 786 svn 1187109463 M * bXi made a script which creates a new gentoo vserver 1187109504 M * nanonyme nice 1187109562 M * nanonyme going to make it public? some other people might like the script too 1187109581 M * bere kernel 2.6.22.2 and vanilla kernel 2.6.20+patch 2.6.21+2.6.22+2.6.22.2....what a different ?? 1187109602 M * bXi nanonyme: http://www.gentoo.org/proj/en/vps/vserver-howto.xml#doc_chap3 1187109612 M * bXi i made a script out of that command :P 1187109616 M * nanonyme ah 1187109643 M * nanonyme sigh, quite a burpy feeling after half a liter of cider... 1187109643 M * Bertl bere: doesn't matter if 2.6.20+patches up to 2.6.22.x or the full version 1187109674 M * bere so if i try on 2.6.22.2 with fb and vserver and my system will hung...this mean bug? 1187109690 M * bXi bere: you should try without the vserver 1187109694 M * bXi and with the vserver patch 1187109712 M * bXi if it happens with a vanilla 2.6.22.2 then its a kernel related bug 1187109723 M * Bertl but keep the same .config, just do a 'make oldconfig' 1187109730 M * bere bxi -without vserver fb work good 1187109732 M * bXi if it doesnt happen in vanilla 2.6.22.2 BUT it does happen in vserver 2.6.22.2 then its a vserver bug 1187109745 M * bXi bere: you said you tested this on 2.6.18 1187109752 M * bere yes 1187109756 M * Bertl exactly, and in this case, it will be fixed asap 1187109766 J * DavidS ~david@85.125.165.34 1187109771 M * bere i not know what "asap" ) 1187109772 M * bXi bere: first try on a vanilla 2.6.22 1187109777 N * DavidS DavidS|Vienna 1187109783 M * bere without vserver? 1187109783 M * bXi your comparing windows 95 to xp so to speak 1187109789 M * bXi without vserver 1187109792 M * bere ok 1187109825 M * Bertl bere: quite simple, get 2.6.22.1 (or 22.2) build it with the config which made fb fail for you 1187109835 M * bXi hmmm 1187109836 M * bXi this is weird 1187109836 M * Bertl bere: then make sure it still fails :) 1187109847 M * bXi Bertl: is this command correct http://www.gentoo.org/proj/en/vps/vserver-howto.xml#doc_chap3 ? 1187109897 M * nanonyme hmm 1187109907 M * Bertl bXi: looks reasonably fine, daniel_hozac? 1187109908 M * bere bertl - on 2.6.22.1 i check 1187109915 M * bere now i will check on 2.6.22.2 1187109935 M * Bertl bere: so you have tried with 2.6.22.1 _without_ the vserver patch? 1187109945 M * bere yes 1187109948 M * Bertl bere: and that works fine for you? 1187109956 M * bere on 2.6.22.1 without vserver fb work good 1187109961 M * bere what? 1187109961 M * nanonyme can i remove su and sudo altogether and only use "vserver enter" to get root privileges on the vserver? 1187109975 M * Bertl bere: and when you add the vserver patch, the kernel hangs? (with the same config)? 1187109983 M * bere yes 1187109986 M * daniel_hozac nanonyme: yes. 1187110003 M * bere hang when try initialize fb 1187110007 M * nanonyme nice. it should reduce security risks quite a lot if no user on the system can become superuser 1187110014 M * bXi weird 1187110015 J * bzed ~bzed@dslb-084-059-114-074.pools.arcor-ip.net 1187110043 M * nanonyme and it's impossible to ssh remote root 1187110047 M * Bertl bere: do you know how to enable kernel debugging? 1187110048 M * daniel_hozac bXi: Hollow keeps the HOWTO updated, looks fine to me. 1187110055 M * bere bertl - gdb? 1187110057 J * Julius ~julius@p57B2797E.dip.t-dialin.net 1187110077 M * nanonyme daniel_hozac: does the command require root has a working password, btw? 1187110081 M * bXi daniel_hozac: from what i remember the last time i made a vserver is that it emerges something with the vserver command 1187110085 M * bXi it didnt do that this time 1187110086 M * Bertl bere: yes, that is an option, but probably you want to start with a few kernel config changes first 1187110086 M * daniel_hozac nanonyme: no. 1187110087 Q * Julius 1187110092 M * nanonyme seriously cool 1187110093 J * Julius ~julius@p57B2797E.dip.t-dialin.net 1187110131 M * Bertl bere: like enabling the NMI watchdog and CONFIG_KERNEL_DEBUG as well as early printk 1187110142 M * bere bertl - maybe i not right understand you (my english bad) but for every kernel i write config ftom null 1187110156 M * bere from* 1187110173 M * bXi in this case you shouldve copied the kernel config from the failing kernel 1187110182 M * bere ok 1187110185 M * Bertl hmm, so you did build two separate configs, one for the unpatched kernel and one for the vserver patched one, yes? 1187110195 M * bere yes 1187110204 M * Bertl bere: I would suggest to do exactly the following: 1187110226 M * Bertl bere: get the 2.6.22.1 kernel source (either via patches or download) 1187110248 M * Bertl bere: copy the .config file from the kernel tree which fails (vserver patched) to this tree 1187110262 M * Bertl bere: run 'make oldconfig' and 'make' 1187110266 M * bere ok 1187110276 M * Bertl try to boot that kernel .. 1187110288 M * Bertl if that hangs too, we know it is a config/kernel issue 1187110338 M * bXi Bertl: got a spare moment? 1187110350 M * bere can do 2.6.22.2 + patch-vserver-2.2.0.3 (form 2.6.22.1 kernel) ?? 1187110367 M * bere or i can download latest version vstable ? 1187110405 M * Bertl bere: you can do all of those, but you _should_ try with the same kernel version which did hang for you :) 1187110462 M * Bertl bXi: what's the problem 1187110769 Q * Johnnie Ping timeout: 480 seconds 1187111223 Q * bere Ping timeout: 480 seconds 1187111434 M * Bertl daniel_hozac: where do I check for things like /bin/nice (path) in the tool config? 1187111492 J * Johnnie ~jdlewis@c-67-163-142-234.hsd1.ct.comcast.net 1187112336 Q * click Ping timeout: 480 seconds 1187112618 Q * dna Quit: Verlassend 1187113320 M * daniel_hozac Bertl: the path the utils are configured with, or to add another tool 1187113321 M * daniel_hozac ? 1187113795 Q * Risjh Quit: Leaving 1187114209 M * Bertl daniel_hozac: I'm just trying to figure which path for example for nice is configured 1187114224 M * Bertl daniel_hozac: we did that the last time, remember? 1187114240 M * daniel_hozac /util-vserver/util-vserver-vars 1187114249 M * Bertl ah, tx 1187114313 M * Bertl venice util-vserver-0.30.214-pre2585 # which nice 1187114313 M * Bertl /usr/bin/nice 1187114318 M * Bertl grep nice ./scripts/util-vserve 1187114318 M * Bertl r-vars 1187114318 M * Bertl _NICE="/bin/nice" 1187114327 M * bXi Bertl: you can do ^a F in the screen btw 1187114336 M * Bertl any ideas? 1187114345 M * bXi will probably give you some more space 1187114378 M * daniel_hozac Bertl: ls -l /usr/bin/nice /bin/nice 1187114390 M * Bertl ls -l /usr/bin/nice /bin/nice 1187114390 M * Bertl ls: cannot access /bin/nice: No such file or directory 1187114391 M * Bertl -rwxr-xr-x 1 root root 20568 Aug 12 17:16 /usr/bin/nice 1187114405 M * daniel_hozac is this on a recently updated Gentoo-box? 1187114410 M * bXi yes 1187114423 M * daniel_hozac did you update coreutils after installing the utils? 1187114428 M * daniel_hozac Gentoo moved nice. 1187114442 M * Bertl I did a configure run right now 1187114457 M * Bertl didn't change the pathes though 1187114465 M * bXi its util-vserver-0.30.214-pre2585 1187114469 M * daniel_hozac in a clean tree? 1187114478 M * Bertl no 1187114481 M * daniel_hozac i.e. no configuration cache? 1187114501 M * Bertl let's see 1187114601 M * DavidS|Vienna gah! runit's debian package tries to install a inittab entry but of course this is not really working. any hints how to workaround that? 1187114694 M * cehteh using rc-file and not runit :) 1187114699 M * Bertl daniel_hozac: okay, problem between monitor and chair ... 1187114747 M * DavidS|Vienna ah, there's already a debian bug about that: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=430951 1187115029 M * eyck DavidS|Vienna: create /etc/init.d/runit file, and changes runit's state to 'installed' in /var/lib/dpkg/status 1187115079 M * eyck runit's author is debian runit maintainer and believes that he has received god's blessing to install stuff in inittab 1187115142 M * DavidS|Vienna i'd need it for git-daemon-run, but i can just as well run git-daemon by hand and ignore the whole shmer 1187115169 M * DavidS|Vienna e.g from inetd 1187115189 M * eyck hmm, runit really is a nice replacement for daemon tools, and it's usefull 1187115230 M * eyck it's just that it's author is.. hmm, we have this term 'beton', for someone as mentaly agile 1187115253 M * DavidS|Vienna hehe, i know "beton" 1187115278 M * eyck there are multiple bug reports about this, dating 4 years back.. and every few months someone re-discovers this, check http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=234106 1187116110 N * DoberMann[PullA] DoberMann 1187116189 Q * bored2sleep Ping timeout: 480 seconds 1187116588 J * bonbons ~bonbons@2001:960:7ab:0:20b:5dff:fec7:6b33 1187117563 Q * matti Ping timeout: 480 seconds 1187118329 J * Piet_ ~piet@tor.noreply.org 1187118765 Q * Piet Ping timeout: 480 seconds 1187118848 J * matti matti@acrux.romke.net 1187119155 J * coderanger_ ~coderange@1cc-dhcp-120.media.mit.edu 1187119859 Q * sladen Ping timeout: 480 seconds 1187119905 J * sladen paul@starsky.19inch.net 1187120235 J * arachnist arachnist@088156185052.who.vectranet.pl 1187120574 Q * gerrit Ping timeout: 480 seconds 1187120758 Q * Piet_ Remote host closed the connection 1187122584 N * _mcp mcp 1187123634 J * click click@ti511110a080-5168.bb.online.no 1187124072 J * Piet_ ~piet@tor.noreply.org 1187124186 N * Piet_ Piet 1187124530 N * DoberMann DoberMann[PullA] 1187124591 M * bXi is Hollow from berlin? 1187124630 M * daniel_hozac he's living there now, IIRC. 1187124639 M * daniel_hozac university. 1187124657 M * bXi ah 1187124688 M * bXi he didnt clean the /etc/resolv.conf in the amd64 stage4 tarball i think 1187125112 M * Supaplex moo 1187125120 M * sid3windr se 1187125360 M * Bertl daniel_hozac: got a minute? 1187126045 J * Aiken ~james@ppp121-45-255-55.lns2.bne4.internode.on.net 1187126068 Q * Aiken Remote host closed the connection 1187126076 J * Aiken ~james@ppp121-45-255-55.lns2.bne4.internode.on.net 1187126746 Q * bonbons Quit: Leaving 1187126799 J * dna ~dna@47-248-dsl.kielnet.net 1187127663 J * bored2sleep ~bored2sle@66.111.53.150 1187128338 Q * Piet Read error: Connection reset by peer 1187128467 Q * DavidS|Vienna Quit: Leaving. 1187128497 J * Piet ~piet@tor.noreply.org 1187128552 J * virtuoso_ ~s0t0na@pppoe-220.58.110.89-adsl.spbnit.ru 1187128960 Q * virtuoso Ping timeout: 480 seconds 1187129151 Q * dna Quit: Verlassend 1187129550 Q * rgl Quit: Zzzzz 1187129678 M * daniel_hozac Bertl: pong 1187129914 Q * meandtheshel1 Quit: Leaving. 1187129920 M * Bertl daniel_hozac: the thing is the following: 1187129945 M * Bertl I would like to make the filesystem tagging more independant from the context ids 1187129953 M * daniel_hozac a did? 1187129963 M * daniel_hozac (disk id :)) 1187129965 M * Bertl (we already talked about that some time ago) 1187129982 M * Bertl yeah, basically I call it 'tag' or 'dlx' 1187129990 M * daniel_hozac dlx? 1187129998 M * daniel_hozac disk limit xid? 1187130031 M * Bertl yeah, although I used tag for all recent patches 1187130038 M * daniel_hozac right. 1187130040 M * Bertl so probably 'tag' it is :) 1187130043 M * daniel_hozac :) 1187130058 M * Bertl now, I see three options there 1187130108 M * Bertl 1) add a tag to each vx_info and make dx_*tag take an indirection over the context (adds some overhead and requires locking) 1187130145 M * Bertl 2) add a tag to each task_struct and set that from a 'default' from nx_info 1187130167 M * Bertl 3) handle it like a separate context, with migrate operations and such 1187130183 M * daniel_hozac s/nx_info/vx_info/, i guess. 1187130193 M * Bertl yeah, sorry 1187130207 M * daniel_hozac 3 seems like a whole lot of work when it's just one variable. 1187130250 M * Bertl well, we wouldn't have to actually 'create' and 'destroy' those contexts 1187130261 M * daniel_hozac yeah, that's true. 1187130265 M * Bertl i.e. they could be 'assumed' just to exist 1187130299 M * daniel_hozac i guess i'm sort of inclined towards 2. 1187130344 M * daniel_hozac though that would mean updating the vx_info wouldn't have any effect on already running processes, which is bad. 1187130350 M * Bertl okay, the only problem I see with that (leaning also towards 2 :) is that we might easily get into inconsistancies 1187130367 M * daniel_hozac right 1187130411 M * daniel_hozac 1 seems like the Right Thing(tm), i'm just not sure it's worth the overhead. 1187130424 M * daniel_hozac 3 seems like the mainline approach :) 1187130515 M * daniel_hozac how about an ugly mix of 1 and 2? 1187130520 M * Bertl btw, is vc_task_xid() actually used inside a context nowadays? 1187130539 M * Bertl or can that be considered legacy :) 1187130553 M * daniel_hozac hmm, only during SETUP, if at all. 1187130580 M * Bertl so we could move the checks out of it, I guess 1187130599 M * Bertl (permission checks that is) 1187130601 M * daniel_hozac see what breaks, at least :) 1187130619 M * Bertl hehe, sounds good to me :) 1187130675 M * Bertl let's hear about your ugly mix ... 1187130742 M * daniel_hozac well, the more i think about it, the more i hate it :) having a pointer to the id (in the vx_info) in the task_struct would avoid the locking (right?), but not the overhead. 1187130781 M * Bertl no, actually that would not help with the locking either 1187130781 M * daniel_hozac hmm, actually, why would 1 need locking at all? 1187130819 M * Bertl hmm, good question, we probably do not access it for other tasks, do we? 1187130828 M * daniel_hozac i don't see why we would. 1187130972 J * DoberMann_ ~james@AToulouse-156-1-150-239.w90-30.abo.wanadoo.fr 1187131065 M * Bertl okay, another question is, if we make that a task property (2 and 3) do we gain/lose something? 1187131080 M * daniel_hozac instant update. 1187131100 M * daniel_hozac and the potential inconsistencies. 1187131122 M * Bertl well, for 3 those are both non issues 1187131123 Q * DoberMann[PullA] Ping timeout: 480 seconds 1187131145 M * Bertl instant update is then on a per process basis 1187131145 M * daniel_hozac yeah, that's true. 1187131239 M * Bertl okay, have to think about that ... 1187131250 M * Bertl (some more) thanks for the time 1187131258 M * Bertl bbl ... 1187131262 N * Bertl Bertl_oO 1187131278 M * daniel_hozac cya! 1187131309 Q * Julius Remote host closed the connection 1187131922 J * Piet_ ~piet@tor.noreply.org 1187131945 Q * Piet Ping timeout: 480 seconds 1187132245 Q * AStorm Remote host closed the connection 1187132269 J * AStorm ~astralsto@tor-irc.dnsbl.oftc.net 1187132441 J * lilalinux_ ~plasma@dslb-084-058-205-209.pools.arcor-ip.net 1187132873 Q * lilalinux__ Ping timeout: 480 seconds 1187133522 Q * arachnist Quit: brb/bbl 1187133692 J * arachnist arachnist@088156185052.who.vectranet.pl 1187134966 J * Linus ~Linus@bl7-132-35.dsl.telepac.pt 1187135486 J * friendly12345 ~friendly@ppp121-44-237-56.lns2.mel4.internode.on.net