1150502583 J * bon bon@193.93.72.10
1150503662 J * serving ~serving@86.108.78.29
1150503875 Q * serving Quit: 
1150503909 J * serving ~serving@86.108.78.29
1150503918 Q * serving Quit: 
1150503932 J * serving ~serving@86.108.78.29
1150504104 Q * serving Quit: 
1150504745 J * sukria_ ~sukria@www.sukria.net
1150504823 J * micah_ ~micah@208.99.202.72
1150504850 Q * sukria Ping timeout: 480 seconds
1150504877 J * Vudu ~vudumen@perverz.hu
1150504900 Q * micah Ping timeout: 480 seconds
1150504901 M * Bertl wb Vudu!
1150504985 Q * Vudumen Ping timeout: 480 seconds
1150508310 J * Kurt kmw@evvlinlwt-nas-07-s199.cinergycom.net
1150509199 M * Bertl welcome Kurt!
1150509347 M * Kurt howdy
1150509361 M * Kurt you would all do well to familiarize yourselves with the writings of the eminent 20th-century Russian-American philosopher Ayn Rand
1150509402 M * Bertl how so?
1150509842 M * doener maybe this one? trying to prove that GPL is capitalistic? http://en.wikipedia.org/wiki/User:TakuyaMurata/Ayn_Rand_and_the_open_source_movement
1150510642 M * cehteh re
1150510782 M * doener wb
1150512401 Q * lilalinux Remote host closed the connection
1150514901 M * Bertl k, I'm off for today ... have a good one everyone! cya tomorrow!
1150514908 N * Bertl Bertl_zZ
1150515090 M * doener same for me... good night!
1150515091 Q * doener Quit: leaving
1150518061 N * micah_ micah
1150518390 Q * FireEgl helium.oftc.net nobelium.oftc.net
1150519154 Q * derjohn2 Ping timeout: 480 seconds
1150519156 J * derjohn2 ~aj@dslb-084-058-204-145.pools.arcor-ip.net
1150519651 J * FireEgl Atlantica@Atlantica.US
1150519794 J * f_ ~f_@83-215-237-1.seek.stat.salzburg-online.at
1150519942 Q * f_ Quit: 
1150525639 J * dna ~naucki@dialer-156-173.kielnet.net
1150525996 J * Kurt2 kmw@216.135.25.156
1150526249 Q * Kurt Ping timeout: 480 seconds
1150527001 Q * dna Quit: Verlassend
1150528210 J * meandtheshell ~markus@85-124-232-137.work.xdsl-line.inode.at
1150528370 J * zkbrsnie ~zkbrsnie@83.64.146.226
1150528899 J * f_ ~f_@83-215-237-1.seek.stat.salzburg-online.at
1150529329 Q * Aiken Ping timeout: 480 seconds
1150531535 J * Viper0482 ~Viper0482@p5497516F.dip.t-dialin.net
1150531621 J * bonbons ~bonbons@83.222.39.166
1150532328 J * Wenix_ ~wenix@81.7.189.11
1150532331 Q * Wenix_ Quit: 
1150534963 Q * f_ Quit: This computer has gone to sleep
1150535538 M * Loki|muh micah: is it possible to add a "mkdir -p /var/run/vservers /var/run/vservers.rev /var/run/vshelper" to /etc/init.d/util-vserver? That would cover that /var/run is a tempfs which it is in ubuntu since dapper. Would be great!
1150536571 J * yelifu ~hongdanst@202.38.114.129
1150537785 J * f_ ~f_@83.215.237.1
1150539953 J * s0undt3ch ~s0undt3ch@bl7-244-107.dsl.telepac.pt
1150544587 J * Kurt kmw@evvlinlwt-nas-07-s254.cinergycom.net
1150544891 Q * Kurt2 Ping timeout: 480 seconds
1150546196 Q * matled Remote host closed the connection
1150547730 Q * lilo2 Quit: leaving
1150547781 J * lilo2 ~0710AAD4@tor-irc.dnsbl.oftc.net
1150550625 N * Bertl_zZ Bertl
1150550629 M * Bertl morning folks!
1150550662 M * Kurt Bertl: got milk?
1150550677 J * orionpanda ~nospam@netblock-66-245-252-180.dslextreme.com
1150550731 M * Bertl Kurt: yes, a big bottle full of ...
1150550736 M * Bertl welcome orionpanda!
1150550748 M * Kurt you know, milk looks like semen
1150550755 M * Kurt so
1150550755 M * Kurt if you pour it over your mouth
1150550756 M * Kurt it looks like you just gave head
1150550761 M * Kurt just something to keep in mind
1150550786 M * Bertl Kurt: I don't pour milk, I usually drink it, but YMMV :)
1150550851 M * Bertl Kurt: btw, do you have any Linux-VServer related questions too?
1150550919 M * Kurt not offhand
1150550925 M * Kurt just hanging out because I'm interested in it
1150550979 M * Bertl i.c. well keep hanging then :)
1150550989 J * Piet ~piet@tor-irc.dnsbl.oftc.net
1150550998 M * Bertl welcome Piet!
1150551002 M * Hollow hey Bertl!
1150551025 M * Bertl hey Hollow! everything fine?
1150551035 M * Hollow yeah, had a nice week in egypt :)
1150551072 M * Hollow i started learning x86 assembler..
1150551084 M * Hollow so confusing :p
1150551087 M * Bertl ah, and I thought visiting pyramids and such ...
1150551106 M * Hollow well, we stayed at the red sea, nothing beside the rsort and desert
1150551127 M * Hollow http://benedikt.boehm.name/gallery/v/Egypt2006
1150551155 M * Hollow phreak told me there will be 128 IPs per ctx without overhead soon?
1150551194 M * Kurt hmm?
1150551196 M * Piet hi bertl
1150551203 M * Kurt are you referring to IPv6?
1150551205 M * Hollow no
1150551206 M * Kurt if so, this is nothing new
1150551206 M * bonbons Hollow: there could probably be 19356 with same overhead ;)
1150551213 M * Hollow ic.. ;)
1150551255 M * Hollow bonbons: how is ipv6 going?
1150551299 M * bonbons it's working for standard udp/tcp (unicast), pings are possible as well
1150551327 M * Hollow sounds good, unfortunately i didn't have any chance yet to play with ipv6
1150551343 M * bonbons just undefined behavior if some IPv6 sockets play IPv4 game
1150551449 M * bonbons thus "echo 1 > /proc/sys/net/ipv6/bindv6only" is recommended
1150551484 M * Hollow i'm not even sure i can buy IPv6 adresses at my isp
1150551517 M * bonbons you can get some from freenet6 (hexago) for free, just need to build tunnel to them
1150551566 M * bonbons that's what I'm doing here as well. No ISP offers IPv6 here... last mile does no allow native IPv6
1150551613 M * bonbons if that was supported I would already have native IPv6 - other problem is support from "normal" ADSL routers
1150551637 M * Hollow well, i'm talking about my dedicated machines here, but hetzner does not offer ipv6 it seems
1150551649 M * bonbons Hollow: for howto look at Gentoo's IPv6 howto ;)
1150551656 M * Hollow heh
1150551664 M * bonbons you could get a tunnel for those as well
1150551690 M * bonbons as those boxes have static IPs other brokers are usable too
1150551714 M * Hollow i.e. the ipv6 connection is tunneld thourhg ipv4?
1150551734 A * Hollow puts on his lamp to see the keyboard better
1150551782 M * bonbons yep, either through UDP+tun or sit tunnel
1150551804 M * orionpanda quick question: which of the two COW fixes (fix05, fix06) should I apply to 2.6.16.20-vs2.1.1-rc22 ?
1150551835 M * daniel_hozac both.
1150551843 M * daniel_hozac they are incremental.
1150551859 M * orionpanda ok. thank you.
1150551939 M * Bertl orionpanda: ocfs2?
1150552003 M * orionpanda yes. i tried the june 12 patch. What's new?
1150552038 Q * Kurt Quit: Leaving
1150552040 M * Bertl http://vserver.13thfloor.at/Experimental/patch-2.6.16-ocfs2-0.02.diff
1150552051 M * Bertl (ext2 attributes for mainline)
1150552083 M * Bertl will soon have a patch with full functionality for vserver
1150552095 M * orionpanda wow. you added full attributes support to ocfs2?
1150552116 M * orionpanda that's fantastic. I'll give this a try.
1150552131 M * Bertl doesn't include the vserver specific ones though
1150552166 M * orionpanda yes. I don't see barrier or iunlink.
1150552194 M * Bertl but, I will make the vserver specific changes _ontop_ of this one
1150552211 M * orionpanda oh. so, this is a patch to OCFS2.
1150552224 M * Bertl yup
1150552243 M * orionpanda makes sense. ocfs2 developers must be thrilled.
1150552272 M * orionpanda question: are there any cache-coherency issues with setting attributes? Or, does ocfs2 not use client-side caching?
1150552302 M * Bertl the patch above is the result of a longer talk with the ocfs2 folks
1150552324 M * Bertl it should work fine across systems (but that's what I would like to see tested :)
1150552419 M * orionpanda right. My setup consists of AOE on each of three machines.
1150552439 M * Bertl should be sufficient I guess
1150552459 M * Bertl the transport medium between systems should not have any relevance
1150552517 M * orionpanda right -- looks like a plain block device. Apparently the performance issues with file copy are ocfs2-specifc. i found a few previous posts.
1150552585 M * orionpanda ocfs2 is not optimized to copy lots of little files. The same issue plagues lustre, too.
1150552846 J * yarihm ~yarihm@80-218-2-149.dclient.hispeed.ch
1150552938 M * Bertl wb yarihm!
1150553050 M * orionpanda is "yarihm" Turkish for "help"?
1150553149 M * orionpanda oops. "help" is "yardim". yarim (no h) means "half". my bad.
1150553404 M * yarihm Bertl: hi
1150553500 M * yarihm orionpanda: ah .. i was told that yarim was aramaeic for friend, i knew it meant something in turkish but I thought it was "sheep" because i once saw an offer for half a sheep from a turkish butcher and then thought the "yarim" was the sheep :)
1150555465 Q * h01ger Read error: Connection reset by peer
1150555495 Q * derjohn Read error: Connection reset by peer
1150555506 J * derjohn ~derjohn@80.69.37.19
1150555688 J * h01ger ~holger@217.114.79.23
1150555732 M * Bertl wb derjohn! h01ger!
1150555775 M * derjohn *lol* it should read: wb derjohn's firewall ;)
1150555796 M * h01ger hi Bertl 
1150555979 Q * yelifu Quit: 
1150558824 J * PilatomiK ~tek@86.218.241.184
1150559129 Q * derjohn Read error: Connection reset by peer
1150559169 J * derjohn ~derjohn@80.69.37.19
1150559241 M * Bertl wb derjohn's fw!
1150559244 M * micah Loki|muh: I'd be happy to add it, can you file a wishlist bug so I can remember?
1150559253 M * derjohn Bertl, heh ;)
1150559452 Q * s0undt3ch Quit: leaving
1150560049 J * s0undt3ch ~s0undt3ch@bl7-244-107.dsl.telepac.pt
1150560248 Q * orionpanda Ping timeout: 480 seconds
1150560784 Q * Snow-Man Ping timeout: 480 seconds
1150560916 J * Snow-Man ~sfrost@kenobi.snowman.net
1150563803 M * PilatomiK salut salut
1150563832 M * PilatomiK y'a t'il des francophone par ici ?
1150563851 M * PilatomiK je suis en train de mettre en place le vserver mais j'ai qq questions
1150563925 M * Bertl no idea what you are talking about, pal! :)
1150563961 M * PilatomiK Bertl I prefere speak french because I'm bad in english
1150563969 M * PilatomiK but I can try in english
1150563994 M * Bertl otherwise you have to look for a 'french' channel
1150564120 M * PilatomiK Bertl french channel must exist ... I'm not sure
1150564171 M * Bertl well, this is an english channel, so we all speak english here (helps us to understand eachother)
1150564308 M * PilatomiK Bertl : I'm on a ubuntu dapper, and I want a sarge vserver. I've compiled kernel from linux-vserver.org with patch with success, and util-vserver
1150564328 M * PilatomiK I create successfully a vserver
1150564332 M * Bertl sounds good, testme.sh works fine?
1150564337 M * PilatomiK but I can't start it
1150564390 M * Bertl how did you build it, and what do you do to start it, and most important, what issue do you encounter?
1150564402 M * PilatomiK vserver-info says VS-API: 0x000100ff
1150564422 M * Bertl looks like the compat/legacy version
1150564432 M * PilatomiK I don't found testme.sh
1150564454 M * Bertl http://vserver.13thfloor.at/Stuff/SCRIPT/
1150564478 M * Bertl upload the output to paste.linux-vserver.org please
1150564487 M * PilatomiK yep :)
1150564597 M * PilatomiK http://paste.linux-vserver.org/93
1150564605 M * Bertl tx
1150564624 M * PilatomiK testme say that "chbind: kernel does not provide network virtualization"
1150564636 M * Bertl first: Use dietlibc: no (you have been warned)
1150564658 M * Bertl second, your kernel has legacy enabled (not unusual for stable)
1150564668 M * Bertl but your tools have the legacy stuff disabled
1150564669 M * PilatomiK but I have compiled the kernel with it normally
1150564689 M * Bertl I'd suggest the following:
1150564708 M * Bertl - install dietlibc
1150564735 M * Bertl - re ./configure the tools with 
1150564735 M * PilatomiK yep
1150564759 M * PilatomiK dietlibc binary or dev package ? I suppose binary 
1150564770 M * Bertl if you have a dev one then this one
1150564782 M * Bertl (usually there is only one package)
1150564816 M * PilatomiK I have installed dietlib and dietlib-dev packages :)
1150564825 M * Bertl ./configure --enable-apis=NOLEGACY
1150564845 M * Bertl you might (in this process) also want to move the prefix
1150564851 M * Bertl i.e. do something like:
1150564864 M * Bertl ./configure --enable-apis=NOLEGACY --prefix= 
1150564900 M * PilatomiK --prefix=/VSERVER for example (I have execute 'mkdir /VSERVER' )
1150564910 Q * yarihm iridium.oftc.net nobelium.oftc.net
1150564910 Q * Piet iridium.oftc.net nobelium.oftc.net
1150564928 M * Bertl well, yes but that will move all the binaries and libraries to /VSERVER which is probably not what you want
1150564950 M * Bertl (i.e. you _will_ have to adjust PATH and LD_PATH)
1150564974 Q * f_ Quit: This computer has gone to sleep
1150564980 M * Bertl what you might want to do is to change the --with-vrootdir=
1150565005 J * yarihm ~yarihm@80-218-2-149.dclient.hispeed.ch
1150565005 J * Piet ~piet@tor-irc.dnsbl.oftc.net
1150565033 M * PilatomiK ./configure --enable-apis=NOLEGACY --prefix=/etc/vserver --with-vrootdir=/VSERVER
1150565040 M * PilatomiK I'm ok ?
1150565134 Q * Hollow Quit: Konversation terminated!
1150565215 M * PilatomiK I have start it with ./configure --enable-apis=NOLEGACY --prefix= --with-vrootdir=/VSERVER
1150565245 M * Bertl yep that looks good
1150565276 M * PilatomiK make running now
1150565278 M * PilatomiK :)
1150565279 M * Bertl nevertheless, you might want to remove the old version from /usr/local/
1150565287 M * PilatomiK okay
1150565294 J * Hollow ~hollow@home.xnull.de
1150565421 M * PilatomiK delete all in /usr/local/var/run and /usr/local/sbin
1150565492 M * PilatomiK make install now
1150565510 M * PilatomiK make install-distribution
1150565514 M * Bertl k, check out the distro-install too (suggested at the end of the install)
1150565564 M * PilatomiK http://paste.linux-vserver.org/94
1150565574 M * PilatomiK (testme.sh)
1150565719 M * PilatomiK vserver compiere build -n ............... is running now (debootstrap in /VSERVER :) )
1150565748 M * Bertl hmm, I suspect the --enable-apis=NOLEGACY didn't work out as expected
1150565772 M * Bertl maybe the old tools are still in the path? /usr/local/bin or so?
1150565792 M * Bertl (at least it looks like)
1150565842 M * PilatomiK nop, only enemy territory files in /usr/local/bin, I have delete /usr/local/sbin who contained vchbind etc etc
1150565889 M * Bertl well, the tools show only v13,net apis
1150565907 M * Bertl if configured with --enable-apis=NOLEGACY they will show:
1150565922 M * Bertl compat,v11,fscompat,v13,net,oldproc,olduts
1150566021 M * PilatomiK make clean before ./configure maybe
1150566077 M * PilatomiK http://paste.linux-vserver.org/95
1150566190 M * PilatomiK I make
1150566192 M * PilatomiK make clean
1150566210 M * PilatomiK ./configure --enable-apis=NOLEGACY --prefix= --with-vrootdir=/VSERVER
1150566211 M * Bertl this is usally configured by the distribution-install
1150566224 M * PilatomiK now make
1150566240 M * PilatomiK maybe it's ubuntu's probleme because ubuntu is not debian !
1150566253 M * Bertl no, that should work quite fine
1150566266 M * Bertl I'm pretty sure you are suffering from previous installs
1150566291 M * PilatomiK (for example .deb vserver patch doesn't work with .deb kernel sources)
1150566294 M * Bertl it might be worth a try to reconfigure to what you had before and do 'make uninstall'
1150566309 M * PilatomiK okay
1150566320 M * Bertl (if you have a history of what you did, that will definitely clean it out)
1150566332 M * PilatomiK I m make the "make uninstall" and verify if the vserver command exist
1150566377 M * PilatomiK after make uninstall bash say bash: /usr/local/sbin/vserver: Aucun fichier ou répertoire de ce type
1150566384 M * PilatomiK not found
1150566395 M * Bertl ah, see /usr/local
1150566425 M * Bertl btw, you get readable messages with export LANG=C LC_ALL=C
1150566456 M * Bertl (i.e. non localized)
1150566473 M * PilatomiK bash: /usr/local/sbin/vserver: No such file or directory 
1150566491 M * PilatomiK when I type vserver on command line
1150566512 M * PilatomiK no vserver binary in /usr/local/bin and /usr/local/sbin
1150566516 M * Bertl just logout and login again
1150566539 M * PilatomiK open a new xterm ?
1150566540 M * Bertl and install the tools with the --prefix= and apis stuff
1150566545 M * Bertl for example
1150566548 M * PilatomiK okay
1150566552 M * PilatomiK I will come back
1150566558 M * Bertl but a new bash or flushing the cash should be sufficient
1150566571 M * PilatomiK okay, I start a new xterm
1150566635 M * PilatomiK make clean || ./configure --enable-apis=NOLEGACY --prefix= --with-vrootdir=/VSERVER
1150566800 M * PilatomiK make
1150566802 M * PilatomiK make install
1150566807 M * PilatomiK make install-distribution
1150566820 M * PilatomiK and testme.sh is the same
1150566834 M * PilatomiK Ea 0.30.210 273/glibc (S*) <v13,net>
1150566871 M * Bertl hmm, very strange ...
1150566890 M * Bertl when you do the ./configure
1150566900 M * Bertl it shows you a page of config settings at the end
1150566918 M * Bertl could you upload that one and compare it to the vserver-info - SYSINFO output?
1150566937 J * f_ ~f_@83-215-237-1.seek.stat.salzburg-online.at
1150566955 M * PilatomiK okay, I prepare a paste.linux-vserver.org page
1150567087 M * PilatomiK http://paste.linux-vserver.org/96
1150567116 M * PilatomiK Available APIs: compat,v11,fscompat,v13,net,oldproc,olduts for configure
1150567123 M * Bertl okay, those are different versions
1150567136 M * Bertl what does `which vserver-info` give?
1150567150 M * PilatomiK and Available APIs: v13,net
1150567161 M * Bertl your vserver-info is from 19:04 CEST
1150567175 M * Bertl i.e. it is definitely an older one
1150567189 M * PilatomiK /usr/sbin/vserver-info
1150567207 M * Bertl okay, the --prefix= installs in /sbin
1150567226 M * Bertl so you probably had a configure/install run with =/usr too
1150567319 M * PilatomiK yep, I can make ./configure --prefix=/usr || make uninstall and restart the make clean || ./configure --enable-apis=NOLEGACY --prefix= --vrootdir=/VSERVER
1150567360 M * Bertl do that, but the || is not necessarily a good idea, unless you do not mean that literally :)
1150567438 M * PilatomiK I execute one command at a time, no more :)
1150567449 M * Bertl good ...
1150567475 M * PilatomiK but actualy I have always /sbin/vserver-info command
1150567568 M * PilatomiK I'm looking for deb package binary (reinstall and uninstall with purge util-vserver deb package)
1150567595 M * Bertl sarge backports and unstable/testing have good packages
1150567610 M * Bertl I'd assume ubuntu should have some too
1150567670 M * PilatomiK hum ... ../kernel-patch/apply/vserver doesn't work
1150567692 M * PilatomiK patch -p1 < ../kernel-patch/diffs/vserver too
1150567707 M * Bertl well, you are better of with mainline kernels, but I was talking about util-vserver
1150567744 J * _coocoon_ ~coocoon@84.160.102.104
1150567751 M * _coocoon_ evening
1150567760 M * Bertl hey _coocoon_!
1150567767 M * _coocoon_ hey bertl
1150567780 M * PilatomiK hummm !!! yes I have installed vserver with deb      http://ubuntu.uni-klu.ac.at/ubuntu.uniklu/  dapper uniklu-vserver
1150567785 M * PilatomiK hello _coocoon_ 
1150567817 M * Bertl PilatomiK: ah, that might clash with your compile too :)
1150567968 M * PilatomiK I don't know what it's the vserver binary (make uninstall is make, no deb package installed) but it say Available APIs: compat,v11,fscompat,v13,net,oldproc,olduts
1150568006 M * Bertl well, that is some progress, recheck with testme.sh
1150568040 M * PilatomiK yes, all succeded
1150568043 M * PilatomiK :)
1150568071 M * Bertl good!
1150568086 M * Bertl now let's continue where you encountered your issues
1150568165 M * PilatomiK I remake util-vserver or I make vserver vservername build .......
1150568167 M * PilatomiK ?
1150568276 M * PilatomiK I must remake
1150568280 M * PilatomiK util-vserver
1150568285 M * Bertl can't hurt to rebuild the guest
1150568440 M * PilatomiK I have remake and make install || make install-distribution
1150568447 M * Bertl okay, good
1150568449 M * PilatomiK deboot strap is running
1150568465 M * PilatomiK (and ./configure with options too)
1150568504 M * PilatomiK testme.sh successfully
1150568517 M * PilatomiK maybe success start of the first vserver
1150568600 Q * nebuchadnezzar Remote host closed the connection
1150568734 J * nebuchadnezzar ~nebu@82.233.222.74
1150568751 M * PilatomiK root@6tek-2:/usr/src/util-vserver-0.30.210# vserver compiere enter
1150568752 M * PilatomiK mesg: /dev/pts/2: Operation not permitted
1150568752 M * PilatomiK compiere:/#
1150568756 M * PilatomiK yeaaaaaaaaaaaaaaaah :)
1150568792 M * PilatomiK big thanks Bertl :)
1150568797 M * bon bertl is the man ;)
1150568799 M * bon hi everyone
1150568807 M * PilatomiK hello bon
1150568810 M * Bertl PilatomiK: you're welcome!
1150568821 M * Bertl hey bon!
1150568839 M * bon so
1150568851 M * bon the grsec+vserver seems to be a stable companion so far for me
1150568865 M * PilatomiK bon : In french bon = good in english
1150568870 M * bon i know
1150568871 M * bon i am good
1150568872 M * bon indeed
1150568872 M * bon ;]
1150568876 M * PilatomiK ;)
1150568913 Q * f_ Quit: This computer has gone to sleep
1150568971 M * PilatomiK postgresql backport for sarge now :)
1150570130 M * PilatomiK I think I must bind my server (postgresql-8.1 on host bind on my host ip, and postgresql-8.0 bind on my vserver ip)
1150570316 M * Bertl yep
1150570325 M * Bertl all services on the host must be restricted
1150570335 M * Bertl the guest services will be automatically restricted
1150570806 J * f_ ~f_@83-215-237-1.seek.stat.salzburg-online.at
1150571036 M * PilatomiK I have some message in the vserver shell-init: error retrieving current directory: getcwd: cannot access parent directories: Aucun fichier ou répertoire de ce type
1150571372 M * Bertl maybe a permission issue?
1150571384 M * bon it's definitely french
1150571385 M * bon ;]
1150571768 M * PilatomiK yesss bon :)
1150572785 J * mire ~mire@16-166-222-85.COOL.ADSL.VLine.Verat.NET
1150572818 M * Bertl welcome mire!
1150575849 M * glen daniel_hozac: i updated that patch once more, maybe you have better ideas about the problem and how to present it to user. http://cvs.pld-linux.org/cgi-bin/cvsweb.cgi/SOURCES/util-vserver-rpm-fake-resolver-badperm-errorlogging.patch
1150576342 M * daniel_hozac i'm not sure how i feel about a daemon spawned by a force loaded library for temporary usage outputting things.
1150576473 M * Bertl daemonic?
1150576566 M * daniel_hozac hehe.
1150576662 M * daniel_hozac are there really any cases where it's expected that / in the guest has bad permissions?
1150576683 M * Bertl yes, on older debian installs? *G*
1150576831 M * Bertl ah, daniel_hozac, you are coding C++, yes?
1150576841 M * daniel_hozac hmm, not really.
1150576867 M * daniel_hozac why?
1150576868 M * Bertl hmm, okay, so you can not tell me about object ownership rules there?
1150576877 Q * yarihm Quit: Leaving
1150576883 M * Hollow bonbons: i've just requested an IPv6 tunnel at sixxs.net, so i can start testing things soon :)
1150576885 M * daniel_hozac hmm, no, sorry. i'm not even sure what that means :)
1150576905 M * Bertl np, will google for it
1150576931 M * bonbons Hollow: cool :) just hope that rebooting your box wont let the pause your tunnel for some time (don't remember which broker does so)
1150576965 M * Hollow i took sixxs.net because they offer access to a PoP at my isp
1150576987 M * bonbons so delays should be the shortest :)
1150576988 M * Hollow so it should be very little latency
1150576990 M * Hollow :)
1150577056 M * bonbons I should be able to setup my own broker at work if I want, then I would have about 5 hops to it, and there I already have native IPv6 :-)
1150577092 M * Hollow heh, maybe i can also request some subnet for my dedicated box, that would be nice
1150577156 M * bonbons from hexago I have a nice /64 subnet
1150577187 M * Hollow the user menu of sixxs has subnet option, but it's disabled.. maybe i need to contact them first
1150577299 M * bonbons don't know, it's some time ago I asked for my tunnel :) way over a year
1150577939 M * _coocoon_ hollow: hello hollow how is the progress of the xml-rpc client sw
1150577992 M * Hollow _coocoon_: good so far, all methods for configuration, start/stop/restart and vshelper are implemented (i.e. the basic stuff works)
1150578037 M * Hollow although i still need to clean up some parts, and a lot of testing will be required then..
1150578055 M * _coocoon_ hollow: hm maybe i can help with testing if u tell me what u need
1150578093 M * _coocoon_ hollow. shall there be also a webgui implemented or have u implemented it till yet
1150578098 M * Hollow apropos.. Bertl: we start implementing the stats collector, any near plans regarding these?
1150578126 M * Hollow _coocoon_: yes, there should be a webinterface too
1150578141 M * Hollow _coocoon_: http://home.xnull.de/misc/vcd.jpg
1150578159 M * Bertl Hollow: hmm? you mean changes or what?
1150578165 M * Hollow Bertl: especially thinking about the min stuff for limits
1150578167 M * _coocoon_ hollow: looks good
1150578206 M * Bertl Hollow: ah, yes, I'd suggest to have a 'reset' command (syscall) which re-adjusts those values
1150578220 M * Hollow yup, i remembered you talked about this at linuxtag
1150578239 M * Hollow will this make it into 2.{0.2,1.1}?
1150578258 M * _coocoon_ hollow. what is with the copy function if there are templates available
1150578294 M * Bertl well, I think we can add it to 2.1.1 should not be that hard, the min infrastrcture is already there
1150578305 M * Hollow _coocoon_: hm? you can still clone vservers (at least in theory, nothing implemented for this yet)
1150578334 M * Hollow Bertl: ok, great, i would test it too if you like
1150578369 M * Bertl sounds good
1150578387 M * _coocoon_ hollow: also another question is does this only work with vserver-utils or also with util-vserver
1150578396 M * _coocoon_ ?
1150578424 M * Hollow _coocoon_: you can install both in paralell, although i see no reason where anyone wants this (except for devs)
1150578440 M * Bertl guest building
1150578449 M * Hollow hm?
1150578464 M * Hollow you can build guests with the daemon too
1150578479 M * Bertl vserver <name> build -m ... (this works with the daemon?)
1150578484 M * Hollow indeed
1150578487 M * Bertl cool
1150578506 M * _coocoon_ hollow: i have no idea about vserver-utils, i have only worked with util-vserver till now
1150578507 M * Hollow there is only one method currently (by template/tarball)
1150578508 M * Bertl so any feature from util-vserver which is not covered yet?
1150578521 M * Bertl ah, okay, so only basic building works
1150578532 M * Bertl not debootstrap or apt-rpm or yum
1150578600 M * Hollow Bertl: look at http://home.xnull.de/misc/vcd.jpg the plan is to make a template cache (only containing tarballs), and the make some build scripts to create these templates.. maybe we can also do it without the intermediate tarball step as an option
1150578620 M * Hollow but yes, in theory it should work with apt, rpm, yum and all those
1150578626 M * Bertl sounds very OVZ-ish to me
1150578657 M * Hollow well, i like it, and if you can optionally skip the tarball phase all is fine
1150578683 M * Bertl it might be nice for working with templates, and I have nothing against it
1150578697 M * Hollow think of gentoo where it would take to compile at least some packages to setup a vservr directly
1150578703 Q * Viper0482 Quit: one day, i'll find this peer guy and then i'll reset his connection!!
1150578707 M * Bertl but I think the network install is much more flexible and keeps stuff up-to-date
1150578725 M * Bertl but maybe both can be combined
1150578734 M * Hollow yeah, thats my idea too
1150578741 M * Bertl would be sufficient if a 'template' could be created from the net
1150578774 M * Hollow a template is just another build method where one of the other build methods was executed before
1150578780 M * Hollow and tarred
1150578784 M * _coocoon_ i do not understand for me it is the best way to have templates and then create and untar it intio the guest
1150578804 M * _coocoon_ the fastest without too much traffic
1150578828 M * Hollow _coocoon_: well, in case of binary distros you always have up to date stuff on vserver creation with the net approach
1150578839 M * Hollow and traffic doesn't really matter today
1150578859 M * Hollow but cpu time for compilation or such, so the template approach is good too for source distros like gentoo
1150578864 M * Hollow i.e. good to have both :)
1150578930 M * _coocoon_ but without templates u have more diskspace
1150578939 M * _coocoon_ ;-)
1150579027 M * Hollow yeah, best if anyone can decide on his own best build method :)
1150579091 M * _coocoon_ hollow: i think about to use the vserver control daemon, so i wanted to hire an opteron server in the next time, so i also thought about a test phase 
1150579108 M * Hollow oh yay, amd64 test hardware is always appreciated :)
1150579143 M * _coocoon_ so i think this is a way to help to get the best 
1150579190 M * Hollow _coocoon_: i roughly expect the first working release in Q4 (does not mean it will contain all util-vserver features)
1150579214 M * Hollow so until then test the hell out of it ;)
1150579230 M * _coocoon_ hm u mean october
1150579237 M * Hollow there are some preliminary test scripts in the tree
1150579251 M * Hollow _coocoon_: right
1150579366 M * _coocoon_ hollow: yes i have seen it but i have not enough programmer experience and also not so much time to get involved in it because of i am working to start a company
1150579397 M * _coocoon_ hollow: i mean if i have the server ui and other guis who are inetrested can also test on it
1150579427 M * _coocoon_ to get the client to work as an open source project work, 
1150579430 Q * zkbrsnie Quit: 
1150579432 M * Hollow _coocoon_: sounds great, i really lack test hardware
1150579452 M * _coocoon_ i thought u have enough
1150579464 M * Hollow it's such a mess to develop util-vserver and vserver-utils on one machine
1150579494 M * Hollow well, just my desktop here..
1150579504 M * Hollow ibook is mostly runnong osx
1150579514 M * Bertl (what a waste :)
1150579539 M * _coocoon_ but if vserver-utils are working fine like util-vserver than it is no problem to use it i think so
1150579557 M * _coocoon_ -are *is
1150579568 M * Hollow Bertl: i like it ;)
1150579598 M * Hollow _coocoon_: the implemented things work as expected, but it still lacks mayn features
1150579608 M * Hollow also the startup process is ugly
1150579618 M * Hollow not configurable enough still
1150579664 M * _coocoon_ hollow: but why have u started this only for vserver-utils and not first for util-vserver, and then ....
1150579715 M * Hollow ehm..?
1150579729 M * Hollow vserver-utils IS vcd
1150579761 M * Hollow there are only some direct syscall command line wrappers, but beside them, vserver-utils only contains vcd
1150579761 M * _coocoon_ sorry i have no idea 
1150579818 M * _coocoon_ what is vcd
1150579835 M * Hollow vserver control daemon
1150579842 M * _coocoon_ oh aha
1150579861 M * _coocoon_ i thought it was another way but the same as util-vserver
1150579898 M * Hollow well, the 1.0 branch was more like this
1150579923 M * Hollow but with 2.0 the daemon will do all the work
1150579936 M * _coocoon_ ok
1150579968 M * _coocoon_ i think there are a lot of things where i have no knowledge, and it never ends
1150579971 M * Hollow there will probably still be a tool like "vserver <name> <command> ..." but the only thing it does is contacting the daemon
1150579981 M * _coocoon_ ah ok
1150580029 M * Hollow the idea is to have multiple clients, for Web, KDE/Gnome, CLI (e.g. in ncurses too or so) anything you can think of basically
1150580149 M * Hollow it would even be possible to create a flex grammar for a vserver command language usable in scripts... sheer endless possibilities ;)
1150580176 M * Hollow but still one central instance keeps track of all things
1150580185 M * _coocoon_ it will maybe never end so it is good
1150580189 M * _coocoon_ ;-)
1150580192 M * Hollow heh
1150580213 M * Bertl Hollow: http://vserver.13thfloor.at/Experimental/delta-rminmax-feat01.diff
1150580220 M * Hollow wow, fast :)
1150580226 M * Bertl tested with vcmd -i 100 -C VCMD_60_9[0] -d
1150580245 M * Bertl i.e. works for me :)
1150580314 M * Hollow ok, will try it soon
1150580322 M * Bertl k, np
1150580532 M * _coocoon_ so hollow maybe u will think about it how i can help 
1150580558 M * _coocoon_ to get it faster to work
1150580559 M * Hollow _coocoon_: yeah, proper test scripts are on top of my todo list
1150580561 M * _coocoon_ ;-)
1150580586 M * _coocoon_ ok i think in the next week i will change my server to opteron server 
1150580591 M * Hollow Bertl: just to verify.. min limits are not used in any way right now, right?
1150580629 M * Hollow and only some use soft limit
1150580665 M * Bertl min limits aka guarantees are not used yet
1150580687 M * Bertl soft limit actually only applies to rss
1150580698 M * Hollow ok, thanks
1150580706 M * Bertl but we will add some kind of penalties for over soft limits soon
1150580713 M * Bertl (at least it is on my todo)
1150580733 Q * meandtheshell Quit: bye bye ... 
1150580830 M * _coocoon_ hollow: which kind of testscripts
1150580848 M * _coocoon_ hollow: i need some help to understand
1150580871 M * Hollow _coocoon_: scripts to perform automated testing of the daemon
1150580908 M * _coocoon_ hollow. ah ok, so if i have the hw we can implemenmt it here to get it to work
1150580922 M * _coocoon_ and then we can let run the scripts ;-)
1150580950 M * Hollow Bertl: hm, why would one to reset the (configured?) min limit/garantees?
1150580965 M * Hollow _coocoon_: exactly
1150580979 M * _coocoon_ ok so i will contact u then
1150581144 M * _coocoon_ Hollow: so u can think about which kernel, which sw and so on to get it to work
1150581162 M * Hollow _coocoon_: which distro are you comfortable with?
1150581175 M * _coocoon_ debian
1150581180 M * Hollow urhg
1150581180 M * _coocoon_ but also fedora
1150581189 M * Hollow but ok..
1150581189 M * _coocoon_ and suse if u want
1150581194 M * Hollow nonono
1150581195 M * _coocoon_ hehe
1150581196 M * Hollow better debian then
1150581199 M * Hollow :)
1150581230 M * _coocoon_ ok soryy but gentoo i have only experince with vserver guest a little bit
1150581250 M * Hollow it would be such fun on an opteron :p
1150581261 J * dna ~naucki@dialer-174-7.kielnet.net
1150581276 M * Hollow but debian is ok too
1150581286 M * Hollow it's just a very long time i used it
1150581307 M * Hollow 2 years or so
1150581338 M * _coocoon_ ok so we can talk about then
1150581349 M * _coocoon_ i will leave now good night to all
1150581354 M * Hollow night
1150581378 Q * _coocoon_ Quit: KVIrc 3.2.0 'Realia'
1150582434 Q * dna Quit: Verlassend
1150583174 M * Nam hey guys
1150583193 M * daniel_hozac hi
1150583281 M * Nam i'm trying to convert my configuration from the old legacy way, to the new method... I am noticing that there is a /etc/vservers/*/fstab file... it looks a bit different then the one i use for the actual vserver...   is it legacy to have the following entry .... "/dev/hdv1       /       ext3       defaults   1 1" in the /vservers/*/etc/fstab?
1150583315 M * daniel_hozac /vservers/*
1150583320 M * daniel_hozac /etc/fstab doesn't matter.
1150583353 M * Nam the /etc/vservers/*/fstab doesn't have the /dev/hdv1.... i used the vscfg-convert.sh from off the linux-vservers site
1150583403 M * daniel_hozac because it's not a device on the host, it shouldn't be in there.
1150583404 M * Nam so, do I need to put the "/dev/hdv1       /       ext3       defaults   1 1" entry in the /etc/vservers/<vserver name>/fstab?
1150583435 M * daniel_hozac no.
1150583496 M * Nam the only thing that needs to be in there is the one for /proc and for /dev/pts? as which was setup be default from the vscfg-convert.sh script?
1150583511 M * daniel_hozac yep.
1150583561 M * Nam nice, thanks danie
1150583565 M * Nam nice, thanks daniel
1150583800 M * Nam i just tried to start the vserver, and I am getting an error... "The cofigured vshelper '/sbin/vshelper' does not match the 'vshelper' script of the util-vserver package" ... what does this mean exactly
1150583805 M * Nam never herd of vshelper before
1150584061 M * Hollow Bertl: still around?
1150584072 M * Bertl Hollow: yup
1150584085 M * Hollow why would one reset the min limit?
1150584108 M * Bertl to track min values?
1150584155 M * daniel_hozac track? i thought it was a guarantee, not the minimum observed?
1150584155 M * Bertl (I assume you refer to the min/max values not the limit)
1150584165 M * Hollow i thought you configure min limits like soft and hard, so if you reset them to the current value, the original setting will be lost, no?
1150584188 M * daniel_hozac Nam: vshelper is what makes halt/reboot -f inside the guest work. IIRC that error message is very descriptive, offering several solutions.
1150584222 M * Bertl Hollow: I guess you are mixing up the limits with the observed/tracked values here
1150584233 M * Hollow probably, so what does the min field track?
1150584240 M * Hollow if it is not a limit
1150584252 M * Bertl the observed minimum, as the max tracks the observed maximum
1150584269 M * Bertl (since the last reset or since the beginning)
1150584282 M * Hollow so, if i never reset the min always is 0?
1150584290 M * Bertl precisely
1150584304 M * Nam yes... i know it offered a bunch of solutions, but I need to know exactly what it is, what it does, so that I can make the right choice of solution
1150584304 M * Hollow oki, thanks
1150584318 M * Nam it says the first suggestion is to disable vshelper
1150584327 M * Bertl Nam: basically it means that your distro install screwed up
1150584344 M * Nam howso?
1150584349 M * Bertl otherwise the proper script would have been installed in proc
1150584359 M * Bertl and the helper would simply be used by the kernel
1150584373 M * Nam /proc/sys/kernel/vshelper?
1150584377 M * Bertl maybe you forgot to make install-distribution or whatever
1150584390 M * Bertl maybe you installed a package which did not set that
1150584399 M * Nam that is set
1150584413 M * Bertl yes, but it points somewhere where the tools do not like it
1150584419 M * Bertl i.e. to a _different_ binary
1150584428 M * daniel_hozac or a non-existant one.
1150584444 M * Nam ah
1150584449 M * Nam i thought the file was in place
1150584452 M * Nam but it was't
1150584454 M * Nam that would be the problem
1150584455 M * Nam hehe
1150584464 M * Nam it was pointing to /sbin/vshelper
1150584471 M * Nam but no vshelper file is installed
1150584499 M * Bertl as I said :)
1150584822 Q * lilo2 Remote host closed the connection
1150584916 Q * f_ Quit: This computer has gone to sleep
1150585056 J * lilo2 ~0710AAD4@tor-irc.dnsbl.oftc.net
1150585206 J * Aiken ~james@tooax6-205.dialup.optusnet.com.au
1150585312 M * Bertl welcome Aiken!
1150585328 M * Aiken hello
1150585549 M * Nam i'm getting the error "chbind: vc_net_create(): Invalid arguement"   ...   this may have something to do with my network configuration, i'm used to the legacy IPROOT=... anyone got pointers or have an idea what the problem is with this issue?
1150585608 M * Bertl well, what does your config look like?
1150585662 M * Nam in /etc/vservers/<vserver name>/interfaces  I have one folder so far
1150585694 M * Nam the folder "0" has 5 files   bcast, dev, ip,mask, and name
1150585738 M * Bertl what about the context id? 
1150585764 M * Nam bcast/ip/mask are set to the IP i want to use... dev says "eth0" which is the iface which hold the IP currently, and the file name has "xyz0" as I was just testing
1150585770 M * Nam of the vserver?
1150585779 M * Nam 1
1150585792 M * Bertl well, 1 is unfortunate, as it is already used :)
1150585802 M * Nam ah
1150585803 M * Bertl 0 = host, 1 = spectator
1150585812 M * Nam start at 2 then
1150585812 M * Bertl 2-49151 is fine :)
1150585880 M * Nam k, now i'm getting a different error... grr....
1150585911 M * Nam "save_ctxinfo: execv(): No such file or directory"
1150585960 M * Bertl --debug is your friend
1150585977 M * Bertl how was the config built, btw?
1150586018 M * Nam http://linux-vserver.org/Legacy-To-Newstyle-Config
1150586031 M * Nam guess it doesn't work that well
1150586031 M * Nam hehe
1150586055 M * Bertl are you migrating old guests?
1150586067 M * Nam trying to yes
1150586070 Q * Piet Write error: connection closed
1150586077 M * Nam --debug?
1150586095 M * Nam for "vserver <vserver-name> start"
1150586097 M * Bertl vserver --debug <name> start
1150586098 J * Piet ~piet@tor-irc.dnsbl.oftc.net
1150586104 M * Bertl wb Piet!
1150586122 M * Piet admit it, that is a script,right?
1150586135 M * Bertl yep, all wetware scripted :)
1150586148 M * Piet :) but a sympathic one.
1150586181 M * Hollow Bertl: i'm still not sure how the stats collection can benefit from min and reset stuff.. i probably still think in the wrong direction.. when would the collector reset these values?
1150586222 M * Hollow why isn't it enough to read current (and max)?
1150586306 M * Nam here is the output Bertl http://pastebin.ca/66682
1150586442 M * Bertl Hollow: sec phone
1150586672 J * doener ~doener@p5487782F.dip.t-dialin.net
1150586686 M * doener gug
1150586712 M * Hollow hey doener
1150587964 Q * bonbons Quit: Leaving
1150588030 M * Bertl k, back
1150588039 M * Bertl Hollow: the idea is the following:
1150588060 M * Bertl let's assume, you collect statistical data every, hmm, minute
1150588076 M * Bertl for example, the number of processes or the VM used
1150588101 M * Bertl now, you can (of course) read the current value and graph that
1150588126 M * Bertl but, (now) you can also read the min/max and graph all three values
1150588129 M * Nam Bertl: Here is a better output, it only has like 190 lines before, this one has the entire output, http://pastebin.ca/66684
1150588142 M * Bertl tx
1150588170 M * Bertl Hollow: the min will be the minimum ovserved since the last reset, and the max the maximum 
1150588182 M * Nam line 800 is line failing
1150588235 M * Hollow Bertl: i see the user for max, but what to do with min/reset?
1150588240 M * Hollow s/user/use/
1150588252 M * Bertl what use do you see for the max?
1150588294 M * Hollow well, if there was a higher usage in the 59.99 seconds you didn't read the stats, you can get that info from max
1150588319 M * Bertl Nam: does /usr/bin/env exist?
1150588336 M * Bertl Hollow: and for a lower one?
1150588376 M * Nam /etc/vservers/hosting/vdir/usr/bin/env does exist yes
1150588386 M * Bertl on the host
1150588449 M * Nam /usr/bin/env is a link to /bin/env, yes, both exist
1150588450 M * Hollow hm..
1150588480 M * Bertl what if the processes read 100 and the max is 101 ?
1150588487 M * Hollow so i reset after every read?
1150588497 M * Bertl does that say anything about the min in this period?
1150588499 M * Hollow and get the average through min/max
1150588508 M * Bertl you get the variance via min/max
1150588537 M * Bertl the average is not accounted, but you can use the current value
1150588562 M * Hollow yeah, i meant variance, didn't know the right word..
1150588571 M * Bertl and yes, you would reset it after reading the values
1150588595 M * Bertl in the future we probably want to record the interval length too
1150588605 M * Bertl i.e. time in msec since last reset
1150588617 M * Bertl (which is pretty easy)
1150588651 M * Hollow ok, it is again that easy i didn't understand it ;)
1150588684 M * Bertl well, consider you reset it at 1000 secs
1150588701 M * Bertl then you wait 60 seconds, read the values at 1061 secs
1150588716 M * Bertl and reset them again (for whatever reason) at 1065 secs
1150588719 M * Hollow ehm.. i meant the min/reset thing..
1150588740 M * Bertl minmax reset it is
1150588771 M * Bertl what I consider funny is that you accept the usefullness for the max, by question the min :)
1150588781 M * Bertl s/by/but