1107216815 Q * hansi33 Quit: ChatZilla 0.9.61 [Mozilla rv:1.7.5/20041226] 1107216826 J * hansi33 ~hansi33@193.80.53.173 1107216838 Q * hansi33 Quit: 1107217308 J * olli4555 ~olli4555@p548C9CDD.dip.t-dialin.net 1107217336 P * olli4555 1107217446 J * olli4555 ~olli4555@p548C9CDD.dip.t-dialin.net 1107217635 P * olli4555 1107218983 Q * Pazzo Quit: Download Gaim: http://gaim.sourceforge.net/ 1107220572 J * UFOczek ufoczek@hood.openbug.net 1107220729 M * UFOczek hello? 1107220739 M * daniel_hozac hi? 1107220813 M * UFOczek is there any patch for 2.6.10? 1107220855 J * oleben56 ~oleben56@p548C9CDD.dip.t-dialin.net 1107220906 Q * serving_ Ping timeout: 480 seconds 1107220907 M * daniel_hozac yes, both 1.9.3 and 1.9.4-rc4 are available for 2.6.10. 1107220968 M * UFOczek can you paste complete link here? 1107220997 M * daniel_hozac http://vserver.13thfloor.at/Experimental/RC-1.9.4/patch-2.6.10-vs1.9.4-rc4.diff 1107221007 M * UFOczek thank you 1107221127 M * UFOczek :) 1107222032 Q * oleben56 uranium.oftc.net keid.oftc.net 1107222032 Q * _are_ uranium.oftc.net keid.oftc.net 1107222032 Q * DuckKing uranium.oftc.net keid.oftc.net 1107222032 Q * Plug uranium.oftc.net keid.oftc.net 1107222032 Q * matta-laptop uranium.oftc.net keid.oftc.net 1107222032 Q * Seraph uranium.oftc.net keid.oftc.net 1107222032 Q * tchan uranium.oftc.net keid.oftc.net 1107222032 Q * we2by uranium.oftc.net keid.oftc.net 1107222032 Q * Snow-Man uranium.oftc.net keid.oftc.net 1107222032 Q * DaCa uranium.oftc.net keid.oftc.net 1107222032 Q * monrad uranium.oftc.net keid.oftc.net 1107222032 Q * TheSeer uranium.oftc.net keid.oftc.net 1107222032 Q * Zoiah uranium.oftc.net keid.oftc.net 1107222032 Q * lilo uranium.oftc.net keid.oftc.net 1107222032 Q * Loki|muh uranium.oftc.net keid.oftc.net 1107222032 Q * stupidawy uranium.oftc.net keid.oftc.net 1107222032 Q * no_maam uranium.oftc.net keid.oftc.net 1107222032 Q * meebey uranium.oftc.net keid.oftc.net 1107222032 Q * locksy uranium.oftc.net keid.oftc.net 1107222032 Q * mcp uranium.oftc.net keid.oftc.net 1107222032 Q * ats uranium.oftc.net keid.oftc.net 1107222054 J * Loki|muh loki@satanix.de 1107222054 J * oleben56 ~oleben56@p548C9CDD.dip.t-dialin.net 1107222054 J * _are_ foobar@pD95EE94B.dip.t-dialin.net 1107222054 J * matta-laptop ~matta@69.93.28.254 1107222054 J * DuckKing ~Duck@dyn-83-157-133-238.ppp.tiscali.fr 1107222054 J * Plug ~plug@datadot.net 1107222054 J * Seraph kk@projects.verfaction.de 1107222054 J * tchan ~tchan@c-24-13-81-164.client.comcast.net 1107222054 J * we2by ~jinxi@145.116.6.75 1107222054 J * Snow-Man ~sfrost@snowman.net 1107222054 J * DaCa ~danny@mail.limehouse.org 1107222054 J * monrad ~monrad@213083190130.sonofon.dk 1107222054 J * TheSeer ~theseer@border.office.salesemotion.net 1107222054 J * Zoiah Zoiah@matryoshka.zoiah.net 1107222054 J * lilo ~lilo@lilo.usercloak.oftc.net 1107222054 J * stupidawy foo@you.wish.you.were.pimp.olicio.us 1107222054 J * no_maam ~erik@datenzone.de 1107222054 J * meebey meebey@meebey.net 1107222054 J * locksy ~locksy@mrtg.sisgroup.com.au 1107222054 J * mcp ~hightower@81.17.110.148 1107222054 J * ats ~as@lotes.vtu.lt 1107222282 Q * oleben56 Quit: Leaving 1107222525 J * Loki|muh_ loki@satanix.de 1107222628 J * oleben56 ~oleben56@p548C9CDD.dip.t-dialin.net 1107222697 Q * Loki|muh_ Remote host closed the connection 1107222702 Q * Loki|muh Read error: Connection reset by peer 1107222825 J * Loki|muh loki@satanix.de 1107222860 Q * micah Remote host closed the connection 1107222865 J * micah micah@micha.hampshire.edu 1107222868 P * oleben56 1107223195 J * oleben56 ~oleben56@p548C9CDD.dip.t-dialin.net 1107223209 Q * chrish01 Quit: Leaving 1107223468 P * oleben56 1107223903 Q * matta-laptop Quit: Hey! Where'd my controlling terminal go? 1107223975 J * oleben56 ~oleben56@p548C9CDD.dip.t-dialin.net 1107224068 P * oleben56 1107224233 Q * ola Ping timeout: 480 seconds 1107226558 Q * berni_ Ping timeout: 480 seconds 1107226739 J * berni ~berni@2001:1b18:202::2 1107228840 J * click click@dsl-84-161.aal.tiscali.no 1107230226 Q * aba Remote host closed the connection 1107230238 J * aba ~aba@sol.turmzimmer.net 1107231239 J * Hollow_ ~bene@home.xnull.de 1107231239 Q * Hollow Read error: Connection reset by peer 1107231458 Q * aba Ping timeout: 480 seconds 1107232140 J * aba ~aba@sol.turmzimmer.net 1107233233 J * nox- ~nox@c135082.adsl.hansenet.de 1107233578 Q * nox Ping timeout: 480 seconds 1107233593 N * nox- nox 1107235172 J * _Plug ~plug@datadot.net 1107235172 Q * Plug Read error: Connection reset by peer 1107235177 N * _Plug Plug 1107235223 J * _Zoiah Zoiah@matryoshka.zoiah.net 1107235223 Q * Zoiah Read error: Connection reset by peer 1107237543 Q * berni Ping timeout: 480 seconds 1107237815 J * berni ~berni@2001:1b18:202::2 1107241682 J * hansi33 ~hansi33@193.80.53.173 1107241763 Q * monrad Quit: Leaving 1107243930 T * services.oftc.net http://linux-vserver.org/ | latest stable 1.2.10, devel 1.9.{3,4-rc4}, ng8.13 1107244279 J * lars||| ~ledj@213-156-52-105.fastres.net 1107244902 J * wfamy bill@lns-vlq-49-mar-82-251-50-122.adsl.proxad.net 1107245236 M * lars||| morning people 1107245546 N * Bertl_zZ Bertl 1107245552 M * Bertl morning folks! 1107245847 M * lars||| welcome back Bertl! 1107245896 M * Bertl hey lars|||! 1107245963 M * wfamy morning bertl 1107246057 Q * _are_ Quit: Disconnecting 1107246062 M * meebey morning 1107246067 M * wfamy Bertl:the mail13.dump and the config file at http://famy.zapto.org/dump 1107246068 M * meebey time for some samba in a vserver action 1107246160 M * Bertl morning meebey! wfamy! 1107246195 N * Doener|gone Doener 1107246202 M * Doener morning! 1107246211 M * Bertl morning Doener! 1107246350 M * Bertl wfamy: guess your outbound connection on that machine is limited to about 7K/s is that right? 1107246394 M * Doener oh, fun to download those files then :) 1107246416 M * Bertl no problem, just will take some time ... 1107246508 M * Bertl Doener: btw, it seems we somehow 'broke' the SCHED_PRIO functionality ... 1107246522 J * agent ~smith@220.224.10.118 1107246569 M * wfamy if you are interested in a debian sarge from scratch with fakeinit init and qmail install problem 1107246606 M * Bertl Doener: I did this quick-fix yesterday http://vserver.13thfloor.at/Experimental/RC-1.9.4/delta-schedprio-fix01.diff and hansi33 was happy with it ... in my tests id had no effect at all (qemu) ... 1107246616 M * wfamy you can download it but there are no info in it. 1107246903 M * wfamy I just open the server and i have 2 wget dl 1107246945 M * wfamy it is at home and have near 15 Kbit upload 1107246970 M * Doener in which way? 1107247071 M * Bertl IMHO the recalculation is not done when SCHED_HARD isn't set 1107247098 M * Bertl (so the tokens in the TB do not change with SCHED_PRIO but not SCHED_HARD enabled) 1107247131 M * Bertl which is not exactly intentional ... IIRC 1107247164 M * Bertl okay, moving out now, back in about 2-3 hours ... 1107247179 N * Bertl Bertl_oO 1107247697 M * bro hello 1107247779 M * Doener hi bro 1107247923 M * bro any1 could help me with my problem? when stoping vserver and after that if i change it's flags i cant start it anymore ;/ config 2.6.9-vs1.9.3 with 0.30.196 tools 1107248008 J * tchan_ ~tchan@c-24-13-81-164.client.comcast.net 1107248027 Q * wfamy uranium.oftc.net keid.oftc.net 1107248027 Q * Loki|muh uranium.oftc.net keid.oftc.net 1107248027 Q * locksy uranium.oftc.net keid.oftc.net 1107248027 Q * meebey uranium.oftc.net keid.oftc.net 1107248027 Q * no_maam uranium.oftc.net keid.oftc.net 1107248027 Q * stupidawy uranium.oftc.net keid.oftc.net 1107248027 Q * TheSeer uranium.oftc.net keid.oftc.net 1107248027 Q * we2by uranium.oftc.net keid.oftc.net 1107248027 Q * DuckKing uranium.oftc.net keid.oftc.net 1107248027 Q * tchan uranium.oftc.net keid.oftc.net 1107248027 Q * Snow-Man uranium.oftc.net keid.oftc.net 1107248027 Q * ats uranium.oftc.net keid.oftc.net 1107248027 Q * Seraph uranium.oftc.net keid.oftc.net 1107248027 Q * lilo uranium.oftc.net keid.oftc.net 1107248027 Q * DaCa uranium.oftc.net keid.oftc.net 1107248027 Q * mcp uranium.oftc.net keid.oftc.net 1107248036 M * bro in short, ive added sched_hard in config file (and yes, ive added it in kernel too) and now the server isnt starting ;/ 1107248054 M * Doener any error messages? 1107248055 M * bro vcontext: vc_create_context(): File exists 1107248095 J * wfamy bill@lns-vlq-49-mar-82-251-50-122.adsl.proxad.net 1107248095 J * DuckKing ~Duck@dyn-83-157-133-238.ppp.tiscali.fr 1107248095 J * Seraph kk@projects.verfaction.de 1107248095 J * tchan ~tchan@c-24-13-81-164.client.comcast.net 1107248095 J * we2by ~jinxi@145.116.6.75 1107248095 J * Snow-Man ~sfrost@snowman.net 1107248095 J * DaCa ~danny@mail.limehouse.org 1107248095 J * TheSeer ~theseer@border.office.salesemotion.net 1107248095 J * lilo ~lilo@lilo.usercloak.oftc.net 1107248095 J * stupidawy foo@you.wish.you.were.pimp.olicio.us 1107248095 J * no_maam ~erik@datenzone.de 1107248095 J * meebey meebey@meebey.net 1107248095 J * mcp ~hightower@81.17.110.148 1107248095 J * ats ~as@lotes.vtu.lt 1107248097 Q * Snow-Man Read error: Connection reset by peer 1107248100 Q * tchan Read error: Connection reset by peer 1107248104 M * bro any ideas? 1107248109 J * Snow-Man ~sfrost@snowman.net 1107248112 Q * lilo Read error: Connection reset by peer 1107248119 M * Doener what does vserver-stat say about the vserver? 1107248119 Q * meebey Read error: Connection reset by peer 1107248121 J * meebey meebey@meebey.net 1107248122 J * lilo ~lilo@lilo.usercloak.oftc.net 1107248138 M * bro server stopped 1107248150 M * bro i can bring it up ONLY if i restart the box 1107248157 M * Doener so it's not listed? 1107248176 M * bro nop 1107248185 M * Doener ok, what about ls /proc/virtual 1107248196 M * bro shows me all contextes 1107248207 M * Doener is the vserver's context included? 1107248222 M * bro yep 1107248232 M * Doener try mount -o remount /proc 1107248267 M * bro same 1107248273 M * bro remounted but nothing ;/ 1107248316 M * Doener cat /proc/virtual//status 1107248328 M * Doener don't paste that in the channel ;) 1107248334 M * Doener either private or some pastebin 1107248395 M * Doener same for cvirt 1107248398 J * Loki|muh loki@satanix.de 1107248428 M * meebey is SMB nat-safe does someone know? 1107248487 M * Doener bro: chcontext --ctx netstat -nal 1107248511 Q * wfamy Ping timeout: 480 seconds 1107248574 M * Doener hm, where do those refs come from.... (just thinking out loud ;) 1107248608 M * Doener let's try: vnamespace -e cat /proc/mouts 1107248614 M * Doener s/mouts/mounts/ 1107248667 Q * stupidawy Ping timeout: 480 seconds 1107248747 M * Doener ok, it doesn't make too much sense that this would keep the context alive, but let's remove those mounts ;) 1107248748 J * rs rs@ice.aspic.com 1107248774 M * Doener vnamespace -e umount /those/mount/points 1107248793 M * agent can anybody tell me if it is possible to change runlevel inside a vserver 1107248811 M * Doener bro: stop 1107248823 M * bro damn 1107248824 M * bro :> 1107248827 M * bro did it already 1107248832 M * Doener is the dev/pts mount still there? 1107248837 M * bro nop 1107248864 J * stupidawy foo@you.wish.you.were.pimp.olicio.us 1107248869 M * Doener ok, guess i'll get a chance to try that out some time 1107248886 M * Doener just wanted to see if there were ptys left 1107248891 M * bro hmm 1107248900 M * Doener is the context gone now? 1107248942 M * bro in /proc/virtual it's still there 1107248964 M * Doener did the usecnt/refcnt change? 1107249024 M * meebey weee now I starting a vserver module based firewall :) 1107249031 M * bro hmm 1107249040 M * meebey vserver_smb.rules 1107249045 M * bro no, it's still the same, 8 and 6 1107249057 M * meebey that way only firewall are loaded which are required for that vserver 1107249072 M * meebey if I am even better I will parse the vserver config file 1107249079 M * meebey and decide then which rules to load 1107249108 M * Doener hmm... at least that's what i expected 1107249149 M * Doener bro: does vnamespace -e cat /proc/mounts list anything? 1107249239 M * bro well, it lists my host's mounts 1107249240 M * bro hmm 1107249247 M * bro and something weird 1107249256 M * Doener weird? 1107249270 M * bro it shows that too 1107249291 M * bro in the botom of the screen 1107249410 M * bro Doener? 1107249424 M * Doener just lost my nameserver... 1107249486 M * Doener in private i just got the three lines showing your host's mount, nothing weird there 1107249507 M * bro hmm 1107249510 M * Doener brb 1107249516 N * Doener Doener|gone 1107249567 N * Doener|gone Doener 1107249579 M * bro hm? 1107249583 M * bro back already? 1107249610 M * Doener just restarted my router... nameserver still dead :( 1107249623 J * prae ~prae@ezoffice.mandrakesoft.com 1107249635 M * bro ;< 1107249711 M * prae morning ! 1107249719 M * bro morning, prae 1107249720 M * Doener ok, let's try with: chcontext --ctx sleep 60 & 1107249739 M * Doener then: chcontext --ctx 1 cat /proc//mounts 1107249830 M * bro hmm, there's a slight problem, i dont see the pid, i mean i see when it starts, but it's not in /proc 1107249862 P * agent Leaving 1107249863 M * Doener it's only visible in context and context 1, thus the chcontext -ctx 1 in front of the cat ;) 1107249884 M * bro oh 1107249957 M * bro hmm, it shows normal mounts picture :| 1107249973 M * bro without that *weird* thingie i pasted in private 1107250010 M * Doener what exactly did you consider to be weird? 1107250072 Q * prae uranium.oftc.net keid.oftc.net 1107250072 Q * stupidawy uranium.oftc.net keid.oftc.net 1107250072 Q * Snow-Man uranium.oftc.net keid.oftc.net 1107250072 Q * no_maam uranium.oftc.net keid.oftc.net 1107250072 Q * TheSeer uranium.oftc.net keid.oftc.net 1107250072 Q * we2by uranium.oftc.net keid.oftc.net 1107250072 Q * DuckKing uranium.oftc.net keid.oftc.net 1107250072 Q * ats uranium.oftc.net keid.oftc.net 1107250072 Q * Seraph uranium.oftc.net keid.oftc.net 1107250072 Q * DaCa uranium.oftc.net keid.oftc.net 1107250072 Q * mcp uranium.oftc.net keid.oftc.net 1107250127 M * Doener those are the vserver's mounts. they're rbind'ed to / 1107250145 J * prae ~prae@ezoffice.mandrakesoft.com 1107250145 J * stupidawy foo@you.wish.you.were.pimp.olicio.us 1107250145 J * Snow-Man ~sfrost@snowman.net 1107250145 J * DuckKing ~Duck@dyn-83-157-133-238.ppp.tiscali.fr 1107250145 J * Seraph kk@projects.verfaction.de 1107250145 J * we2by ~jinxi@145.116.6.75 1107250145 J * DaCa ~danny@mail.limehouse.org 1107250145 J * TheSeer ~theseer@border.office.salesemotion.net 1107250145 J * no_maam ~erik@datenzone.de 1107250145 J * mcp ~hightower@81.17.110.148 1107250145 J * ats ~as@lotes.vtu.lt 1107250342 M * Doener hm... no processes, no sockets, too many refs for the infamous proc "bug"... 1107250372 M * bro don't scare me ;> 1107250407 M * bro when i restart the server everything will work as planned 1107250445 M * bro but IF i make any changes to vserver's config {flags|schedule} it's all gone 1107250466 M * Doener you did: stop->change config->start? 1107250470 M * bro yep 1107250484 M * Doener legacy config? 1107250496 M * Doener hm, no. wouldn't use namespaces 1107250663 J * wfamy bill@lns-vlq-49-mar-82-251-50-122.adsl.proxad.net 1107250778 M * wfamy Bertl_oO: the installation crash when i do package/install in daemontools installation from http://www.qmailrocks.org/qmail.htm 1107250898 M * bro Doener, by any chance i could succeed with starting the vserver? or there are more things to test? 1107250972 M * Doener i'll have a look at the source now, trying to figure things out... if you don't mind the dangling context, you could just change the vserver's configuration to use an other context 1107251001 M * bro ok 1107251005 M * bro ill try 1107251053 M * bro yep, starts up just fine when i changed the context 1107251071 J * chrish01 ~chrish01@69.90.131.25 1107251314 J * Mega\work ~Megabart2@host56-101.pool80182.interbusiness.it 1107251379 M * Mega\work hello world :D 1107251423 M * chrish01 hi Mega\work 1107251729 M * meebey how do I allow multiple existing network interfaces in a vserver? 1107251739 M * meebey I tried: IPROOT="eth0 eth1" 1107251749 M * meebey just eth0 works 1107251794 M * chrish01 not sure. i would wait fro Bertl_oO to get in 1107251810 M * chrish01 i only use the 1.9 stuff myself 1107251813 M * meebey k 1107251816 M * meebey ic 1107251825 M * Doener http://vserver.13thfloor.at/Stuff/VServer-IP-Setup-0.1.txt might be of help 1107251834 M * meebey checking that out 1107251971 M * meebey hm that doesn't help 1107252058 M * Doener it should at least tell you, that the chbind stuff doesn't know about interfaces 1107252195 M * Doener what do you want to do/have? 1107252210 M * meebey I want to allow interfaces which I dont know the IP for 1107252217 M * meebey like ppp0, dynamic IP 1107252260 M * Doener only possible with 0.0.0.0 or ALL... that's one of the things where ng will strike ;) 1107252268 M * meebey hrm 1107252276 M * meebey thats bad because I often got MANY devices 1107252288 M * meebey the tools complain about too many at 16 already 1107252291 M * meebey 16 is nothing :) 1107252400 M * Doener then 0.0.0.0 is your only choice... ALL uses each ip address seperatly as argument to chbind, while 0.0.0.0 just gives full access IIRC 1107252414 M * meebey sure? 1107252423 M * meebey when I use 0.0.0.0 ifconfig will not show any IP 1107252428 M * meebey any IP of the devices 1107252437 M * Doener try with "ip a" 1107252459 M * meebey ip a? 1107252469 M * meebey vpn_galilei:/# ip a 1107252469 M * meebey 1: lo: mtu 16436 qdisc noqueue 1107252469 M * meebey link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 1107252469 M * meebey 2: eth0: mtu 1500 qdisc pfifo_fast qlen 1000 1107252469 M * meebey link/ether 00:e0:18:84:ee:cc brd ff:ff:ff:ff:ff:ff 1107252471 M * meebey 3: eth1: mtu 1500 qdisc pfifo_fast qlen 1000 1107252474 M * Doener but basically it should set the ipv4root to 0 and disable all checks... 1107252474 M * meebey link/ether 00:02:b3:b6:f7:90 brd ff:ff:ff:ff:ff:ff 1107252487 M * Doener interesting... 1107252487 M * meebey that breaks programs though 1107252499 M * meebey the programs can't detect which IPs they can use 1107252528 M * Doener got to wait for Bertl then, i'm not really familiar with the 2.4 stuff 1107252535 M * meebey ipv4root is now 0.0.0.0 1107252556 M * meebey 2.4 is the only choice for servers, 2.6 is still too buggy :) 1107252603 N * Bertl_oO Bertl 1107252617 M * Bertl meebey: I read that! ;) 1107252640 M * Bertl (but of course you mean the kerne, not linux-vserver, right?) 1107252654 M * Bertl morning folks! 1107252687 J * locksy ~locksy@mrtg.sisgroup.com.au 1107252698 M * Bertl welcome locksy! 1107252750 M * Mega\work morning Bertl 1107252824 M * chrish01 morning Bertl 1107252885 M * Bertl morning Mega\work! chrish01! 1107253011 M * Mega\work How can chroot the vserver dir? 1107253062 M * Mega\work at the moment, i have this permission in a vserver dir 1107253064 M * Mega\work drwxr-xr-x 6 root root 4096 Jan 22 11:19 vservers 1107253086 M * Bertl hmm, are we talking about 2.4 or 2.6 here? 1107253235 J * Duckx ~duckx@195.75.27.158 1107253273 M * Mega\work 2.4 1107253371 M * Bertl okay, you should change it with: chmod 000 /path/to/vserver/.. and chattr +t /path/to/vserver/.. then 1107253384 M * Mega\work good 1107253392 M * Bertl (this will make it a barrier ;) 1107253674 Q * Hollow_ Ping timeout: 480 seconds 1107253675 J * Hollow_ ~bene@home.xnull.de 1107253714 M * meebey Bertl: is there any solution to my problem? 1107253744 M * meebey Bertl: 0.0.0.0 does not work as expected 1107253759 M * meebey ALL does but with the 16 ips limit 1107253798 M * meebey IPROOT="eth eth1 ppp0" more I dont need 1107253805 M * Bertl meebey: if I understood that right, you want no limitation at all, right? 1107253806 M * meebey IPROOT="eth0 eth1 ppp0" even 1107253821 M * meebey Bertl: limit would be ok, but on devicename basis 1107253835 M * Bertl yeah, but you can do without a limit, right? 1107253842 M * meebey yes 1107253857 M * Bertl so what you want is not to use chbind at all ... 1107253877 M * meebey ah not specifying IP*? 1107253879 M * meebey lemme try 1107253891 M * Bertl IIRC the alpha tools disable the chbind if no address is specified 1107253892 Q * Hollow_ Remote host closed the connection 1107253905 M * meebey nope 1107253909 J * Hollow ~bene@home.xnull.de 1107253916 M * meebey ifconfig shows no IPs at all 1107253918 M * meebey only devices 1107253945 M * meebey AFAIK no IPROOT means IPROOT=0.0.0.0 implicit 1107253947 M * Bertl that could be related to the hiding 1107253948 M * Doener Bertl: in kernel/vserver/context.c - vx_migrate_task() we could end up double-putting a vx_info, if the task to be migrated already had a vx_info 1107253951 M * meebey and that gives no IPs 1107253969 M * meebey Bertl: hm 1107253977 M * meebey Bertl: how can I disable the hiding? 1107253993 M * Bertl Doener, meebey: sec 1107254000 Q * Hollow Remote host closed the connection 1107254013 J * Hollow ~bene@home.xnull.de 1107254222 M * Doener Bertl: nvm, is already fixed in rc4 1107254233 J * serving ~serving@213.186.180.153 1107254608 M * Bertl meebey: bug report/feature request, there is no flag which controls the hiding in 2.4 ;) 1107254622 M * Mega\work Bertl: if i had kernel 2.6, what was the command for chroot vserver? 1107254643 M * Bertl meebey: so what you want is to disable the chbind 1107254676 M * Bertl Mega\work: setattr --barrier /path/to/vserver/.. 1107254711 M * Bertl okay, luch time ... back in an hour or so ... 1107254718 N * Bertl Bertl_oO 1107254734 M * Mega\work other question 1107254755 M * meebey Bertl_oO: yes 1107254760 M * Mega\work is there vserver for alpha processor? 1107254776 M * meebey AFAIK vserver is arch independend 1107254779 M * meebey just compile and try 1107254920 M * Mega\work but, i must only compile kernel with patch, and run vserver utils, right? 1107255153 M * wfamy Bertl_oO: tell me when download finish 1107255578 Q * Mega\work Ping timeout: 480 seconds 1107255956 M * Doener bro: still around? 1107256224 M * bro yep 1107256229 M * bro still here 1107256314 M * Doener ok, cat /proc/virtual//limits please 1107256734 M * Doener any chance that you ran out of memory while the vserver was running? 1107257279 M * Doener Bertl_oO: http://doener.homeip.net/doener/vserver/diff-2.6.11-rc2-vs1.9.4-rc4-fix-copy_mm.diff 1107257803 J * Mega\work ~Megabart2@host56-101.pool80182.interbusiness.it 1107257809 M * Mega\work re 1107258399 M * bro Doener, no, it's almost impossible 1107258419 M * Doener ok 1107258730 Q * Mega\work Quit: Leaving 1107258761 J * Mega\work ~Megabart2@host56-101.pool80182.interbusiness.it 1107258785 Q * nox Ping timeout: 480 seconds 1107259312 N * Bertl_oO Bertl 1107259332 J * nox ~vps@c135082.adsl.hansenet.de 1107259379 M * Bertl Doener: hmm, interesting 2.6.11-rc* only or general? 1107259388 M * Doener was in 1.9.3 as well 1107259408 M * Bertl excellent, maybe the long standing context won't go away issue? 1107259427 M * Doener possible, i'm checking all set/clr_vx_info calls atm 1107259450 M * Bertl good, if you are doing that, maybe I should explain where I want to go with 1.9.5 1107259451 M * Doener bro has some context without process/sockets and a use/refcnt of 6/8 1107259479 M * Doener limits still show vm/rss/files usage 1107259488 M * Bertl I'm panning to replace the refcnt/usecnt by a single counter, and keep a task counter 1107259508 M * Bertl basically doing the hash/unhash on first task enter/last task exit 1107259527 M * Bertl thus eliminating all needs to do cleanup from irq or wherever 1107259554 M * Bertl (should fix the preemption issues too) 1107259744 Q * wfamy Quit: My damn controlling terminal disappeared! 1107260064 M * lars||| can I use same configuration for 2.6 kernels as with 2.4? 1107260150 M * Bertl yes, both configs (legacy and newstyle) are supported on both kernels 1107260194 M * lars||| that is vservername.conf ? 1107260196 M * lars||| ok 1107260218 M * Bertl vservername.conf is the legacy config, yes 1107260237 M * lars||| ok, is there an example of the newstyle? 1107260274 M * Bertl http://linux-vserver.org/alpha+util-vserver (look at the flower page) 1107260290 M * lars||| ok, found it, awful page design! 1107260301 M * Bertl which one? 1107260311 M * Loki|muh lars|||: you can switch styles 1107260389 M * lars||| Bertl: the flower page 1107260439 M * lars||| Loki|muh: yeah I think I will use the new style later 1107260472 M * lars||| but is 1.9.4-rc4 working out-of-box with the old style? 1107260478 M * Bertl yes 1107260481 M * lars||| ok 1107260499 J * brian ~brian@pD9563740.dip.t-dialin.net 1107260505 M * brian Hi all 1107260535 M * lars||| May I ask, why information is spread across so many websites... have you considered using a CMS ? 1107260544 M * bro hmm, any1 got any ideas how to stop a vserver(legacy) on 2.6.9 kernel with 30.196 utils and IF i dont have any clue at all what context is this server using 1107260581 M * bro plain vserver blabla stop doesn't work 1107260600 M * bro giving me chcontext: vc_new_s_context(): Operation not permitted 1107260616 M * bro oh, and it used to use dynamic context 1107260624 M * brian when it gives a stable version for kernel 2.6 ?? my sew server has a raid only supported by 2.6 :( 1107260752 M * bro any1? 1107260807 M * brian hey bro, any1 sleep :-) 1107260859 M * bro ;/ 1107260877 M * bro it's like 2.27pm here ;p 1107260904 M * brian here 1.27 pm 1107260981 M * lars||| is it just split to files in newstyle config vs. oldstyle 1 file config? 1107261070 J * agent ~smith@220.224.10.118 1107261103 M * agent hi Bertl 1107261139 M * agent can u please tell me how does proc file system works in a vserver 1107261196 M * lars||| anyway would anyone care to send me an example of a running vserver, just a tar'ed /etc/vservers/vserver-name/ directory 1107261313 M * Bertl agent: well, basically as in every other server ... 1107261335 M * Bertl lars|||: look at the Download section on linux-vserver.org 1107261365 M * agent does every vserver have its own proc 1107261384 M * Bertl nope, the proc is shared ... 1107261416 M * agent this means each vserver can see the information about the other 1107261602 M * brian Hi Bertl 1107261655 M * lars||| I can't find the example, must be hidden :) 1107261692 M * rs re 1107261697 M * rs hey Bertl! 1107261702 M * bro hm, so, bout stopping a legacy vserver with the new tools, is it possible? 1107261895 M * lars||| bro: can't you just use the old tools to stop, and start it with the new tools again? 1107261956 M * bro hmm, havent tought bout it ;| 1107261986 M * bro wouldn't it screw up my current configuration? 1107262096 M * lars||| I don't know 1107262165 M * Bertl hey rs! 1107262169 M * lars||| maybe it should be started by the new tools in order to be stopped by the them too 1107262190 M * bro it was started with the new tools 1107262196 M * bro but the server is legacy type 1107262199 M * bro i cant stop it no more 1107262201 M * lars||| ok 1107262216 M * lars||| weird :) 1107262220 M * bro and since it uses dynamic context it makes reeeealy hard to guess which one is that 1107262223 M * Bertl lars|||: http://www.lycos-vds.com/dists/, http://debian.marlow.dk/vserver/guest 1107262266 M * Bertl agent: no, the processes are isolated between them (this includes the proc view) 1107262290 M * lars||| Bertl: it's not those :) It's the configuration with newstyle 1107262325 J * nish ~nish@220.224.10.118 1107262342 M * nish evening all :) 1107262430 M * agent Bertl: how does that happen if the proc is shared 1107262920 M * Bertl evening nish! 1107262949 M * lars||| quick question: do I need to patch 1.9.4-rc4 to get quota ? 1107262967 M * Bertl agent: the proc is a virtual filesystem, so you can have different views into the same filesystem 1107262982 M * lars||| is there some document describing the configuration of the kernel ? 1107262990 M * Bertl lars|||: 1.9.x doesn't support quota on shared disk only disk limits) 1107262991 Q * agent Read error: Connection reset by peer 1107263007 M * lars||| Bertl: ok 1107263020 Q * nish Ping timeout: 480 seconds 1107263025 M * Bertl lars|||: well, you can get an example config by using the -skeleton build method in a few seconds ... 1107263051 J * agent ~smith@220.224.24.58 1107263098 M * lars||| Bertl: ok thanks 1107263169 M * bro Bertl, could u give me any hints on how to stop a legacy vserver using the new tools(3.196)? 1107263205 M * bro s/3/30/ 1107263323 M * Bertl vserver stop ? 1107263410 M * bro umm, no 1107263420 M * bro mind if i paste in private? 1107263427 M * Bertl go ahead ... 1107263439 J * nish ~nish@220.224.24.58 1107263770 Q * flock Ping timeout: 480 seconds 1107263964 J * flock ~restless@l192-117-111-12.broadband.actcom.net.il 1107264138 M * lars||| is the menu "Linux Vserver" the only place for kernel configuration in kernel 2.6? 1107264154 M * lars||| in 2.4 I seem to remember it was different places...? 1107264167 M * Bertl yep, for now it's this menu only ... 1107264188 M * Bertl (there are a few additional changes, like the split patches) 1107264229 M * Bertl a good idea, if you want to know what stuff a patch changed, is to configure the vanilla kernel, then apply the patch and 'make oldconfig' 1107264250 M * Bertl Doener: http://vserver.13thfloor.at/Experimental/RC-1.9.4/delta-copy_mm-fix01.diff http://vserver.13thfloor.at/Experimental/RC-1.9.4/delta-64bit-fix01.diff 1107264499 M * lars||| what is the split patches? 1107264534 M * Bertl different memory splits for x86 1107264568 M * lars||| ok, should I apply them ? 1107264575 A * lars||| is asking a lot :) 1107264581 M * Bertl they are part of 1.9.4-rc* 1107264666 M * lars||| ok 1107264687 M * lars||| but they are configured elsewhere than "Linux Vserver" then? 1107264700 M * Bertl yes, they have a separate menu point ... 1107264749 M * Bertl as I said, if you want to see all new options, best method is to take a config for the vanilla kernel and check the new stuff with make oldconfig 1107264765 M * lars||| ah, I sse 1107264975 M * nish how to get init to listen to /dev/initctl inside chroot 1107265010 M * Bertl lars|||: and btw, we have a saying here: "One who doesn't ask, stays dumb." 1107265048 M * Bertl nish: as you stated at the other channel, it's communicating over a fifo, right? 1107265059 M * nish yeah 1107265087 M * Bertl so if the init process creates that fifo inside the chroot, telinit can use it ... 1107265120 M * nish hmm.. 1107265228 T * Bertl http://linux-vserver.org/ | latest stable 1.2.10, devel 1.9.{3,4-rc4}, ng8.13 -- He who asks a question is a fool for a minute; he who doesn't ask is a fool for a lifetime 1107265295 M * nish He who asks a question is a fool for a minute ?? not at all ! 1107265348 M * Bertl ;) 1107265398 M * bro hmm 1107265400 M * bro :> 1107265639 M * nish Bertl: i've installed a basesystem on a loopback device and 1107265652 M * nish ive chrooted inside that 1107265683 M * nish now i want to spawn a init inside that chroot environment 1107265700 M * nish ?? 1107265756 M * nish when i just write init 3 ,inside chroot bash prompt ,it gives error 1107265765 M * Bertl will not be too easy, most init require to use pid=1 1107265785 M * Bertl actually telinit and init are often the same binary 1107265805 M * nish saying "timeout opening/wrtiting control channel /dev/initctl" 1107265817 M * nish yeah its just the link to init 1107265826 M * lars||| hehe nice topic 1107265834 M * Bertl nish: yes, because it doesn't have pid=1 it acts as telinit 1107266080 M * nish what if i write sys call and invoke it from user space which will just fork init.( from sched_init onwards) 1107266115 M * Bertl will not change the fact that it will not get pid=1 1107266233 M * nish ok lets forget that it wont get pid=1 (we will fake it for desecendent processes) 1107266276 M * nish but still would it work as the original init works 1107266289 M * Bertl where is the point in doing a special syscall then? well of course you can do it this way ... 1107266405 M * lars||| what should I use in "Persistent Inode Context Tagging" in the kernel ? 1107266426 M * Bertl basically the default 24/24 should be fine ... 1107266433 M * lars||| ok 1107266500 M * lars||| "Enable Hard CPU Limits" can be disabled with newstyle configuration, right? 1107266517 M * Bertl not related 1107266559 M * Bertl the hard scheduler will allow you to specify a hard limit for the cpu usage per timeintervall via the token bucket 1107266587 M * Bertl (commonly referred to as %cpu ;) 1107266596 M * lars||| ok 1107266607 M * lars||| is this done in the kernel also? 1107266621 M * Bertl yes, the token bucket has to work in the kernel ... 1107266643 M * Bertl the configuration is done from userspace of course 1107266779 M * lars||| and userspace is from the configuration for each vserver? 1107266804 M * Bertl huh? 1107266823 M * lars||| hehe 1107266839 M * lars||| what is userspace? 1107266853 M * Bertl userspace is where your applications run .. 1107266864 M * Bertl your shell, the syslog, etc ... 1107266864 M * lars||| ok 1107266886 M * Bertl kernel space is the palce where the kernel (and kernel threads) reside 1107266911 M * lars||| but can I configure each vserver with a CPU hard limit in its own configuration? 1107266929 M * Bertl yes, every vserver has a separate token bucket ... 1107266961 M * lars||| ok, now I'am with you... so i'm going to use ulimits on the host server? 1107266985 M * Bertl ulimits are something which applies to processes 1107267012 M * Bertl so they are just valid for one specific process, but every process can have different ones 1107267030 M * Bertl there are rlimits per context, which limit the amount of memory or file descriptors 1107267038 M * Bertl for the entire context 1107267068 M * Bertl and there are scheduler properties, mostly configured via the token bucket for the process scheduling inside a vserver 1107267092 M * lars||| ahh I see 1107267109 M * lars||| so I'm going with rlimits when enabling Hard CPU limits 1107267164 M * Bertl hmm, no, actually rlimits are always enabled regardless of the hard cpu scheduler 1107267218 M * Bertl which reminds me that the current schedprio fix is insufficient ... 1107267240 M * lars||| ok 1107267425 M * lars||| then I don't understand the Hard CPU limit in kernel config if rlimit can be used to limit an entire context... 1107267435 Q * Mega\work Ping timeout: 480 seconds 1107267454 M * Bertl lars|||: you know what rlimits there are? 1107267484 M * Bertl man setrlimit 1107267540 M * Bertl what limits would you set if you would like a vserver to use 25% of your system's cpu power? 1107267669 M * lars||| mmmhh 1107267695 M * lars||| I don't see a limit of the cpu, only seconds etc. 1107267716 M * Bertl precisely ... that is why we added the token bucket (scheduler) 1107267766 J * Mega\work ~Megabart2@host56-101.pool80182.interbusiness.it 1107267771 M * Bertl and it is designed to either act as 'guide' adjusting the priorities for the processes (in SCHED_PRIO mode) or to enforce the assigned shares (in SCHED_HARD mode) 1107267793 M * lars||| ok, so it's done automatically? 1107267804 M * lars||| so 4 vservers will do 25 % each? 1107267816 M * lars||| or approximately 1107267846 M * Bertl if you configure it properly, yes 1107267927 M * lars||| well, now you mention configure... :) Where is this done? Is it just the overall processes and ressource consumption etc. inside a context which should equal for each vserver ? 1107267931 J * pulsar ~pulsar@82.209.246.234 1107267959 M * Bertl welcome pulsar! 1107267966 M * pulsar Hi! 1107267981 M * Bertl lars|||: http://linux-vserver.org/Linux-VServer-Paper-06 (section 06.3) 1107268140 J * mhepp ~mhepp@r72s22p13.home.nbox.cz 1107268182 M * pulsar 2.6.9-vs1.9.3 randomly hangs on SMT vwuth HT, unfortunatly server is co-located, so can't grab ooops message .. 1107268213 M * Bertl welcome mhepp! 1107268224 M * Bertl pulsar: hmm, could you try 1.9.4-rc4? 1107268273 M * mhepp Hello! 1107268284 M * pulsar I think reinstal kernel on 20:00 GMT 1107268329 M * Bertl okay, check what's recent then, probably 1.9.4-rc5 will be out by then 1107268543 M * lars||| Ok, i'm gonna breathe some fresh air... see you later 1107268555 A * lars||| is away: gone breathing 1107268693 Q * nish Remote host closed the connection 1107268712 M * Bertl cya 1107268906 J * nish ~nish@220.224.24.58 1107268922 M * Bertl wb nish! 1107268933 M * nish Hi 1107269035 M * Bertl mugwump: you around? 1107269094 M * Bertl Doener: somehow the vavavoom calculation looks dubious to me ... 1107269240 M * matti Bertl: ;] 1107269249 M * Bertl hey matti! 1107269450 Q * nox Ping timeout: 480 seconds 1107269547 N * chrish01 chris|breakfast 1107270723 Q * mhepp Quit: mhepp caught signal: Autobus error 1107271006 J * nox ~vps@c135082.adsl.hansenet.de 1107271280 Q * Mega\work Ping timeout: 480 seconds 1107271392 M * Bertl welcome nox! 1107271400 M * nox thanks Bertl 1107271484 M * nox had a froze at 1.2.10 , but dunno find a reason 1107271497 M * nox forward still work but no in/out 1107271507 M * nox worked 1107271510 M * Bertl hmm, please elaborate? 1107271546 M * nox Bertl: i'm still searching 1107271577 M * nox but had no more ssh nor connection to any vps 1107271616 M * nox maybe hardwarestuff, the machine is old 1107271640 M * Bertl could be ... no serial console? 1107271672 M * nox no i'm sorry 1107271715 J * Megabart ~Megabart2@host56-101.pool80182.interbusiness.it 1107271720 M * Megabart hello world! 1107271728 M * Bertl hello Megabart! 1107271733 Q * lars||| Ping timeout: 480 seconds 1107272128 M * Megabart :D 1107272607 M * micah good morning 1107272618 M * Bertl morning micah! 1107272710 M * micah Bertl: I saw enrico's response about the tools todo list and saw he mentions "vcopy" -- I did not know this exists and I have been trying to figure out what it does 1107272775 Q * agent Ping timeout: 480 seconds 1107272775 M * Bertl well, I'm clueless here as well ... 1107272778 M * micah it almost looks like the vserver new build -m skeleton; cp -a /vservers/reference/* /vservers/new replication method 1107272787 M * micah but there are no docs anywhere : 1107272788 M * micah :9 1107272790 Q * nish Ping timeout: 480 seconds 1107272795 J * agent ~smith@220.224.27.76 1107272818 J * nish ~nish@220.224.27.76 1107273099 M * Bertl hansi33: ahh, a new day and the mystery unveils ;) 1107273110 M * Bertl micah: use the source! 1107273137 M * hansi33 Bertl: what was the mystery? 1107273151 M * Bertl I used the wrong tool to set the flag ;) 1107273163 M * Bertl (i.e. it was not set at all ;) 1107273234 M * hansi33 Bertl, welcome to the club 1107273243 M * Bertl but as a side effect the scheduler did get a nice cleanup ;) 1107273264 M * Bertl and thanks for spotting the issue! 1107273270 M * hansi33 do you have another patch now? 1107273299 M * Bertl it will be in 1.9.4-rc5, but if you like, you can test it right now ... 1107273315 M * hansi33 it was the first time to solve such problems with irc, most of the time google was my friend. 1107273359 M * Bertl yeah, well, irc has the interactivity bonus ;) 1107273403 M * hansi33 if i "speak" with irc to you, i don't who fast you are typing. I think its faster than my thinking. 1107273413 M * hansi33 who->how 1107273423 J * lars||| ~ledj@213-156-52-105.fastres.net 1107273442 M * Bertl wb lars|||! 1107273544 M * Bertl hansi33: http://vserver.13thfloor.at/Experimental/RC-1.9.4/delta-schedprio-fix02.diff (you have to revert the previous first) 1107273686 M * Bertl Doener: (if you get around to have a look at it) http://vserver.13thfloor.at/Experimental/RC-1.9.4/delta-schedprio-fix02.diff 1107273722 A * Doener yawns 1107273730 M * Doener hm, fell asleep... 1107273772 M * Bertl IMHO the sleep in the morning is the best ... 1107273883 M * Doener yep, and sleeping in the afternoon makes me feel sleepy the rest of the day :( 1107273982 M * lars||| what is the delta patch for ? Hard CPU limit? 1107274501 M * Bertl nope, it fixes an issue with the SCHED_PRIO and cleans up the vavavoom calculation 1107274593 N * chris|breakfast chrish01 1107274608 M * Bertl wb chrish01! enjoyed your breakfast? 1107274614 M * chrish01 no dude! 1107274616 M * chrish01 it sucked ass 1107274621 M * nish :) 1107274633 M * chrish01 some fucker hit my door (which i have a $5,000 paint job) 1107274640 M * chrish01 and then he tried to deny it 1107274647 M * hansi33 Bertl, the patch does not work for 1.9.3, because i dont't have include/linux/vserver/sched_def.h 1107274647 A * nish fell from bed laughing 1107274662 M * chrish01 total moron. i love proving old people wrong 1107274669 M * chrish01 s/proving/prooving 1107274688 M * Bertl hansi33: ah, yes, right, well, either go for the 1.9.4 then or leave it as it is ... 1107274728 M * Bertl chrish01: I guess you are talking about your 'car' door, right? 1107274771 M * chrish01 yea 1107274772 M * chrish01 =P 1107274824 M * Bertl well, I'm sorry for you then ... this probably spoiled your day (somewhat) 1107274836 M * chrish01 totally 1107275132 M * lars||| bbl 1107275132 Q * Megabart Read error: Connection reset by peer 1107275164 Q * lars||| Quit: using sirc version 2.211+KSIRC/1.3.11 1107275238 J * Pazzo ~thomas@host130-250.pool8172.interbusiness.it 1107275402 M * Bertl welcome Pazzo! 1107275421 M * Pazzo hi bertl! 1107275444 M * Pazzo *grrrrrrrrrrmbl* #*+"ยง$ 1107275454 Q * brian Quit: using sirc version 2.211+KSIRC/1.3.11 1107275464 M * Bertl hmm, had a bad breakfast time too? 1107275484 M * Pazzo bertl: linux, specially with kernel 2.6 may be really, really buggy... but I WILL NOT WILL NOT WILL NOT switch back to windows 1107275501 M * Pazzo no, had a short night and a long day :-( 1107275510 M * UFOczek Bertl: :) 1107275564 M * Pazzo installed a windows host, really careful, with all patches, service pack etc - firewall, antivirus, firefox & thunderbird, antispy and all the little things you need to get a more or less useable system 1107275595 M * Bertl UFOczek: long not seen! 1107275601 N * nish nish|dinner 1107275649 M * Pazzo then spent two hours trying different printers for an old sharp printer (drivers suggested by the reseller) - with a great result: windows is unbootable. no way. totally messed up :( 1107275667 M * Pazzo I SPENT A WHOLE DAY DOING ABSOLUTELY NOTHING 1107275681 N * Pazzo Spaccato 1107275699 M * Bertl well, you already said you spent it with windows ;) 1107275826 M * Spaccato customer service :-( 1107275828 N * Spaccato Pazzo 1107275892 J * DuckMaster ~Duck@dyn-83-157-140-234.ppp.tiscali.fr 1107275978 M * Bertl a friend of mine once said, as I asked him why he sells the M$ stuff: hey, if I sell them a linux box as server, it's just working, and I'm out of business, so I have to driver there every week, and they pay for their server management ... 1107276062 M * Bertl (was some time ago, M$ stuff probably requiress less maintainence these days ;) 1107276150 M * Doener if you don't account the anti-virus/malware/spyware updates and stuff ;) 1107276259 A * Bertl notes that he should take a look at debugfs ... 1107276326 Q * DuckKing Ping timeout: 480 seconds 1107276600 M * Pazzo Doener (hi!): only Warmduscher's install updates, antivirus and such stuff! 1107276631 M * Pazzo (Warmduscher, Schattenparker, Servicepackinstallierer!) 1107276637 M * Doener ;) 1107276682 M * Bertl transl. "Warmduscher": mollycoddle, wimp 1107277160 M * chrish01 Bertl, i will have money from my insurance company by noon (in 3 hours) :) 1107277236 Q * pulsar Quit: Leaving 1107277237 M * Bertl well, good ;) 1107277243 M * chrish01 talk about fast claim adjustments haha 1107277253 M * chrish01 so maybe my day wont suck quite so bad 1107277255 M * Doener yep, debugfs looks really interesting 1107277473 M * Bertl seems the assert_lock patches have benn accepted ... 1107277608 M * Bertl (in mm2) 1107277735 M * Doener i'll take a shower, maybe that gets me awake... back later 1107277738 N * Doener Doener|gone 1107278205 P * agent Leaving 1107278566 Q * prae Quit: Client exiting 1107278606 N * nish|dinner nish 1107280301 M * micah is it possible to allow a specific file access from one vserver to another? like allowing vserver2 to symlink into vserver1:/var/www/data ? 1107280324 M * micah by setting maybe an attribute on vserver1:/var/www/data to make it accessible outside of its context? 1107280396 M * micah I know it kinda breaks the context concept, but it is a very limited break 1107280440 M * daniel_hozac hardlinks should work. 1107280452 N * Doener|gone Doener 1107280455 M * Pazzo micah: or mount --rbind 1107280556 M * micah daniel_hozac: but the files in /var/www/data change names and do so frequently, even if i did hardlinks every 1 minute it would not be good 1107280570 M * micah Pazzo: this is an interesting idea, I do not know --rbind 1107280625 M * Pazzo man mount => "Since Linux 2.4.0 it is possible to remount part of the file hierarchy somewhere else. The call is 'mount --bind olddir newdir'" 1107280629 M * micah Pazzo: what is the 'r' in --rbind? 1107280632 M * Pazzo mount --rbind is coooool :-) 1107280655 M * daniel_hozac --rbind is to remount all filesystems mounted under olddir at newdir as well. 1107280665 M * Pazzo yep 1107280701 M * Pazzo micah: you need to do the mount before the vserver starts - otherwise it will not be visible in the other namespace 1107280719 M * Pazzo (at least was so with 1.9.3, didn't retry with 1.9.4-rc4) 1107280721 M * micah daniel_hozac: ah, so the 'r' is just the remount 1107280739 M * daniel_hozac Pazzo: or do the mount in the namespace ;) 1107280776 M * Pazzo micah: no -> man mount 1107280787 M * micah daniel_hozac: wouldn't I have to do the --rbind in the host system? 1107280796 M * daniel_hozac micah: you would. 1107280811 M * Pazzo the "r" probably stands for recursive (my guess) as it makes also submounts visible 1107280814 M * micah Pazzo: man mount doesn't describe --rmount it only describes --mount 1107280824 M * micah err 1107280832 M * micah Pazzo: man mount doesn't describe --rbind it only describes --bind 1107280862 M * Pazzo daniel_hozac: doing the mount in the namespace is a good idea... I've never tried it out, thnx! 1107280865 M * daniel_hozac my man mount describes it ;) 1107280871 M * Pazzo my one too :- 1107280873 M * Pazzo ) 1107280878 M * micah how do you mean doing the mount in the namespace? 1107280899 M * Doener Pazzo, daniel_hozac: or put it into the vserver's fstab if you want it all the time ;) 1107280904 M * Pazzo vnamespace --help 1107280926 M * daniel_hozac Doener: of course, but that's too easy :) 1107280930 M * Doener ... that's the fstab in the vserver configuration, not the vserver's /etc 1107280941 M * Doener *g* 1107280978 M * micah Pazzo: vnamespace --help only tells me the options, but not at all what vnamespace *does* 1107281016 M * micah Doener: how can you put a mount in the vserver's fstab that is in a different context? 1107281044 M * Pazzo vnamespace enters/creates/... namespaces - you can probably find docs about namespaces on www.linux-vserver.org 1107281068 M * micah Doener: ooh in the vserver config, so the vserver's config's fstab gets executed on the host system 1107281144 M * Doener micah: yes. and the host's whole mount tree is still available in the vserver's namespace when the vserver is running, it's just hidden from the vserver. otherwise the vnamespace trick wouldn't work ;) 1107281161 M * micah Doener: I dont understand the vnamespace trick yet :) 1107281195 M * Doener vnamespace just puts your current process into the vserver's namespace, thus you can modify its mount tree 1107281214 M * Doener s/current/the chosen/ 1107281274 M * DaCa Doener: what happens with open files that are outside that tree? 1107281275 M * aba Can somebody give me a hard cluebat about IPv6 status? Is the status with the current development version: (a) no IPv6 at all, (b) no IPv6 in the clients, and does using the ngnet patches change anything? 1107281276 M * micah oh, so if vserver2 was xid=42, you would do vnamespace -n -e 42 mount /vservers/vserver1/var/www /var/www 1107281303 M * daniel_hozac or just vnamespace -n -e vserver2... ;) 1107281328 M * micah daniel_hozac: yes, but I want this vserver to always have this mount point, so probably I want to set it in the vserver config's fstab 1107281370 M * daniel_hozac micah: right, i was just saying that the tools take a vserver name as well as an xid most everywhere. 1107281390 M * micah ah, ok, good to know, the vnamespace --help tool doesn't mention that 1107281439 M * Doener DaCa: the process is started in that namespace, the open files should be stdin/out/err, so no problem there AFAICT 1107281497 M * micah hmm 1107281522 M * micah vnamespace -n -e vserver2 --mount --rbind /home /home gives me the error: 1107281526 M * micah Can not specify multiple operations; try '--help' for more information 1107281529 M * DaCa Doener: oh, I misunderstood then, I thought it would move a running process into another mount tree 1107281573 M * Doener no, "chosen process" was still at least as bad as "current process", i know ;) 1107281614 M * Doener back later... 1107281618 N * Doener Doener|gone 1107281635 M * micah hmm, the fstab entry is not working either 1107281655 M * micah oh wait, maybe... 1107281689 M * micah hmm 1107281715 M * micah I put this in fstab: 1107281723 M * micah none /home ext3 defaults 0 0 1107281736 M * micah and it tells me: /usr/local/etc/vservers/MT/fstab:4:1: failed to mount fstab-entry 1107281746 M * Doener|gone micah: needs to have "bind" as options 1107281751 M * micah ahhh 1107281765 M * micah or rbind 1107281766 M * Doener|gone and "none" is probably not what you want ;) 1107281786 M * micah maybe I'll just try with /home 1107281811 M * micah yeah, that works :) 1107281814 M * Doener|gone gotta be like: /what/to/mount /where/to/mount auto bind 0 0 1107281823 M * micah s/bind/rbind 1107281833 M * Doener|gone depends on your needs ;) 1107281837 M * micah right :) 1107281837 M * micah thanks 1107281841 M * micah it works 1107281849 M * Doener|gone you're welcome... really gone now ;) 1107281852 M * micah ;) 1107282260 N * Bertl Bertl_oO 1107282639 M * micah how can I add an IP address to a vserver? do something to /etc/vservers//interfaces? 1107282752 M * micah it used to be /etc/vservers/.conf IPROOT= 1107282762 M * aba micah: yes. 1107282768 M * aba interface/0/* 1107282778 M * aba (and 1/*, and 2/*, ...) 1107282785 M * micah aba: so I mkdir -p interface/0 1107282799 M * micah but then what is the format of interface/0? 1107282801 M * aba micah: no. 1107282823 M * aba micah: My hint is: Create one testvserver with the appropriate tools, and transfer the configuration from there. 1107282953 M * micah ah, its easy 1107282954 M * micah thanks 1107283037 M * micah hmm, when I restart my vserver I get this error: RTNETLINK answers: Cannot assign requested address 1107283355 J * pulsar ~pulsar@213.184.238.34 1107285491 Q * Duckx Quit: Leaving 1107285864 Q * rs Quit: home 1107285873 J * Megabart Megabart@host232-71.pool80181.interbusiness.it 1107285877 M * Megabart hi all 1107285989 M * pulsar hi 1107286301 M * micah i set a vserver to have an IP 69.12.133.221, the host has .220... when I enter the vserver eth0 doesn't have that IP configured, and the host responds to the .221, whats going on? 1107286356 M * micah yes, I restarted the vserver after I gave it that IP 1107286473 J * prae ~prae@sherpadown.net 1107286479 M * pulsar micah> stable or devel branch ? 1107286524 J * wfamy bill@lns-vlq-49-mar-82-251-50-122.adsl.proxad.net 1107286590 M * wfamy hi 1107286619 M * micah pulsar: alpha/devel branch 1107286776 M * pulsar micah> do U do echo 'IP' > /usr/local/vservers/etc/vservers//interfaces/0/ip 1107286813 M * micah pulsar: yes 1107287042 M * micah pulsar: except that it is /usr/local/etc/vservers//interfaces/0/ip ) 1107287148 M * pulsar path to vserver configs direcory 1107287171 M * micah yes, but that 0/ip file has the IP address 1107287340 M * micah maybe it was because I did vserver test build -m skeleton --interface 69.12.133.221 --netdev eth0 1107287346 M * micah first... then I deleted that vserver 1107287351 M * micah and then created a new one? 1107287691 M * micah ah 1107287702 M * micah its because my sshd and apache on the host are binding to 0.0.0.0 1107287719 M * pulsar ;-) 1107287725 M * micah good to figure out 1107287853 A * mugwump is in the house 1107292906 Q * pulsar Ping timeout: 480 seconds 1107293086 J * pulsar ~pulsar@82.209.237.8 1107293737 N * Bertl_oO Bertl 1107293741 M * Bertl evening folks! 1107293758 M * nish evening Bert ! 1107293769 M * Bertl evening nish! 1107293798 M * aba Hi Bertl 1107293948 J * yarihm ~yarihm@80-218-7-247.dclient.hispeed.ch 1107293962 M * Bertl hello aba! yarihm! 1107294274 M * mugwump Bertl, you were looking for me? 1107294293 M * Bertl yes, I stumbled over your vavavoom code ... 1107294317 M * mugwump go on... 1107294321 M * Bertl and because of the missing comments (I have to add some I guess) I didn't understand it on the first attemp ... 1107294328 M * Bertl +t 1107294345 M * Bertl but after verifying the code once again, it's now all clear to me ... 1107294356 M * Bertl so sorry for the noise ... 1107294367 M * mugwump Just curious, which part did you think was under-commented? 1107294407 M * mugwump effective_vavavoom? 1107294418 M * Bertl http://vserver.13thfloor.at/Experimental/RC-1.9.4/del-vs1.9.4-rc2/14_sched.diff 1107294439 M * Bertl yes the effective_vavavoom() code 1107294667 M * mugwump So, it was unclear why I was multiplying those numbers together etc? The word "quadratically" in the preceding comment is the clue 1107294774 M * Bertl no, that wasn't unclear, it was unclear why the values got higher and higher for lower token values until they reach ... 0 1107294848 M * mugwump that's the scheduler. priority goes from 0 (best) to 140 (worst) 1107294859 M * mugwump as you probably found out by now 1107294874 M * Bertl yes, what confused me was the fact that for token <=0 vavavoom returns 0 1107294882 M * Bertl s/<=/