1251763542 Q * dowdle Remote host closed the connection
1251767952 Q * yepsen 
1251768563 Q * geb Ping timeout: 480 seconds
1251768758 Q * Piet Ping timeout: 480 seconds
1251768822 J * Piet ~piet@659AAB7GU.tor-irc.dnsbl.oftc.net
1251771546 Q * bzed Remote host closed the connection
1251771554 J * bzed ~bzed@devel.recluse.de
1251771898 Q * cehteh Ping timeout: 480 seconds
1251772059 J * cehteh ~ct@pipapo.org
1251774032 J * saulus_ ~saulus@c193183.adsl.hansenet.de
1251774444 Q * SauLus Ping timeout: 480 seconds
1251774450 N * saulus_ SauLus
1251775280 Q * bzed Remote host closed the connection
1251775287 J * bzed ~bzed@devel.recluse.de
1251776321 M * Bertl off to bed now .. have a good one everyone!
1251776325 N * Bertl Bertl_zZ
1251777391 J * geb ~geb@earth.gebura.eu.org
1251780490 Q * geb Ping timeout: 480 seconds
1251780491 J * doener_ ~doener@i59F55A90.versanet.de
1251780589 Q * doener Ping timeout: 480 seconds
1251780674 J * geb ~geb@earth.gebura.eu.org
1251781210 J * nkukard ~nkukard@196.209.246.155
1251782677 Q * Marillion Remote host closed the connection
1251782745 J * Marillion ~dirk@hetzner4.127011.net
1251784458 J * sharkjaw ~gab@90.149.121.45
1251784873 J * davidkarban ~david@199.123.broadband11.iol.cz
1251785325 J * thierryp ~thierry@home.parmentelat.net
1251787097 Q * thierryp Remote host closed the connection
1251787258 Q * aj__ Ping timeout: 480 seconds
1251787924 J * thierryp ~thierry@home.parmentelat.net
1251787949 Q * hparker Quit: Read error: 104 (Peer reset by connection)
1251788155 J * BenG ~bengreen@94-169-110-10.cable.ubr22.aztw.blueyonder.co.uk
1251788305 Q * thierryp Remote host closed the connection
1251788379 J * thierryp ~thierry@home.parmentelat.net
1251789683 J * mxs_ mxs@p4FCCA47D.dip.t-dialin.net
1251789840 J * dna ~dna@55-197-103-86.dynamic.dsl.tng.de
1251789990 Q * mxs Ping timeout: 480 seconds
1251790444 J * hijacker_ ~hijacker@213.91.163.5
1251790444 Q * hijacker Read error: Connection reset by peer
1251790510 J * dna_ ~dna@80-204-103-86.dynamic.dsl.tng.de
1251790589 J * dna__ ~dna@117-201-103-86.dynamic.dsl.tng.de
1251790608 J * derjohn_mob ~aj@80.85.196.112
1251790949 Q * dna Ping timeout: 480 seconds
1251791010 Q * dna_ Ping timeout: 480 seconds
1251791730 Q * derjohn_mob Ping timeout: 480 seconds
1251792608 J * Pazzo ~ugelt@reserved-225136.rol.raiffeisen.net
1251794882 Q * thierryp Remote host closed the connection
1251794935 J * thierryp ~thierry@home.parmentelat.net
1251796787 J * derjohn_mob ~aj@80.85.196.112
1251797451 N * Bertl_zZ Bertl
1251797455 M * Bertl morning folks!
1251797741 M * arekm mording
1251797789 M * arekm Bertl: is there some way to make cpu scheduling equally split between vhosts (way that don't uses that ugly tocken bucket algorithm where user needs to figure out some correct numbers)
1251797797 M * arekm s/vhosts/guests/ of course 8)
1251797854 M * Bertl CFS :)
1251797888 M * Bertl but you still need to figure out 'some correct numbers' :)
1251797965 M * arekm http://linux-vserver.org/util-vserver:Cgroups - these numbers are a lot simpler 8)
1251797999 M * arekm now what 512 is ... aka can it be just 1 without some bad impact? 1, 1, 4, 1 as in that example (512, 512, 2048, 512)
1251798095 M * Bertl should work, keep in mind, that the default (mainline kernel) is 1024 or so ... just that you don't wonder why your guests won't get any scheduling time if some 'other' cgroup is active :)
1251798629 M * geb morning Bertl :)
1251799869 J * balbir ~balbir@59.145.136.1
1251800023 Q * geb Ping timeout: 480 seconds
1251801608 J * ktwilight_ ~keliew@129.10-240-81.adsl-dyn.isp.belgacom.be
1251801905 Q * ktwilight Ping timeout: 480 seconds
1251802183 Q * thierryp Remote host closed the connection
1251803459 M * Bertl nap attack ... bbl
1251803465 N * Bertl Bertl_zZ
1251803846 M * arekm Bertl_zZ: and cpuset.mem does what? 
1251803958 M * arekm and how to view live cgroup stats or something?
1251804066 J * ghislainocfs21 ~Ghislain@LPuteaux-151-41-11-129.w217-128.abo.wanadoo.fr
1251804301 Q * ghislainocfs2 Ping timeout: 480 seconds
1251804458 M * ghislainocfs21 but TB let you put top limits where cpu groups just cannot (if i remember right)
1251804683 M * arekm and figuring out sane tb numbers is a pain
1251805014 J * thierryp ~thierry@home.parmentelat.net
1251805030 Q * balbir Ping timeout: 480 seconds
1251806017 J * hparker ~hparker@2001:470:1f0f:32c:290:96ff:fe50:40fa
1251806091 Q * BenG Quit: I Leave
1251806928 J * SHINSAKU ~Shinsaku@chello089076140236.chello.pl
1251807309 J * BenG ~bengreen@94-169-110-10.cable.ubr22.aztw.blueyonder.co.uk
1251807680 Q * davidkarban Quit: Ex-Chat
1251809843 J * mrfree ~mrfree@host1-89-static.40-88-b.business.telecomitalia.it
1251810469 M * arekm Bertl_zZ: http://pld.pastebin.com/f6ab1d40b
1251811688 Q * transacid Ping timeout: 480 seconds
1251811947 J * transacid ~transacid@transacid.de
1251812155 Q * groente Ping timeout: 480 seconds
1251813022 N * Bertl_zZ Bertl
1251813026 M * Bertl back now ...
1251813137 M * Bertl arekm: could you addr2line the addresses for me?
1251813184 M * arekm can't now but this is some mainline problem I guess. looks similar http://lkml.org/lkml/2009/8/25/210
1251813252 M * Bertl yes, looks like, but the addresses would clarify :)
1251813385 M * Bertl Linux-VServer doesn't change the fair scheduler, but we schedule and unschedule tasks (if the hard cpu limit is used)
1251813601 Q * sharkjaw Remote host closed the connection
1251813697 Q * BenG Quit: I Leave
1251813986 J * dowdle ~dowdle@scott.coe.montana.edu
1251816473 J * bakins ~bakins@157.166.167.129
1251816529 Q * mrfree Quit: Leaving
1251817846 Q * thierryp Remote host closed the connection
1251817863 Q * derjohn_mob Ping timeout: 480 seconds
1251818594 Q * SHINSAKU Read error: Operation timed out
1251819379 J * ghislainocfs2 ~Ghislain@LPuteaux-151-41-11-129.w217-128.abo.wanadoo.fr
1251819610 Q * ghislainocfs21 Ping timeout: 480 seconds
1251820169 Q * Pazzo Quit: Ex-Chat
1251821338 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d
1251821818 J * derjohn_mob ~aj@80.85.196.112
1251823349 J * geb ~geb@194.4.82-79.rev.gaoland.net
1251823428 Q * geb 
1251823489 N * DoberMann[ZZZzzz] DoberMann[PullA]
1251824130 Q * Supaplex Ping timeout: 480 seconds
1251824274 J * hijacker ~hijacker@87-126-142-51.btc-net.bg
1251824659 M * ktwilight_ has the behaviour of apps/init/mark changed recently? when i mark it to default, the guest doesn't start.
1251824689 M * ktwilight_ Kernel: 2.6.27.31-vs2.3.0.36.6
1251824689 M * ktwilight_ util-vserver: 0.30.216-pre2833
1251824706 M * BWare check your os's init script
1251824721 M * ktwilight_ what am i looking for?
1251824725 M * BWare s/os/distro/
1251824739 M * BWare if it uses the apps/init/mark file
1251824924 M * ktwilight_ hm, looks like there's this vservers-default thing in debian that doesn't kick in
1251824926 M * ktwilight_ thanks BWare :)
1251824941 M * BWare you're welcome
1251824947 M * ktwilight_ strange that it works on a different machine though
1251824955 M * ktwilight_ and without vservers-default running
1251824957 M * Bertl well, debian is different :)
1251824959 M * BWare perhaps different init script
1251824975 M * ktwilight_ it looks identical. both deb versions are hte same too
1251824987 M * ktwilight_ some debian weirdness :)
1251824998 M * Bertl debian decided that 'vservers-default' is not needed
1251825019 M * Bertl they somehow combined the runlevel scripts to a single big one
1251825038 M * Bertl (ignoring the fact, that some folks might use marks different form 'default')
1251825046 M * ktwilight_ hm, strange that one of my deb machines works with vservers-default and one without.
1251825052 M * ktwilight_ the machine just booted (with vservers-default) and it works
1251825075 M * Bertl if you install util-vserver from source, it will get the 'usual' 3 runlevel scripts
1251825114 M * ktwilight_ oh! my util-vserver versions are different!
1251825209 M * ktwilight_ oh, i compiled my own util-vserver...no wonder...
1251825212 A * ktwilight_ facepalm
1251825217 M * ktwilight_ thanks Bertl
1251825410 M * Bertl np
1251825631 N * DoberMann[PullA] DoberMann
1251825664 Q * derjohn_mob Ping timeout: 480 seconds
1251829253 J * imcsk8 ~ichavero@nat.ti.uach.mx
1251830225 J * thierryp ~thierry@home.parmentelat.net
1251831013 Q * thierryp Remote host closed the connection
1251831323 J * Snow-Man ~sfrost@tamriel.snowman.net
1251831342 M * Snow-Man How can I mount something on the host and then make it visible to a vserver..?
1251831351 M * Snow-Man --bind, --rbind, etc, don't appear to be working.
1251831406 M * Bertl they work, but are you sure you want to do that and not something slightly different?
1251831429 M * Snow-Man what I'm looking to do is figure out some way to mount a ISO image so the vserver can see it
1251831434 M * Bertl like, e.g. mount something into the guest :)
1251831446 M * Snow-Man when I mount it on the host, it all looks fine
1251831462 M * Snow-Man but then I try to use vnamespace -e blah mount --rbind /mnt /path/to/vserver/mnt
1251831462 M * Bertl yes, and it is mounted within the host namespace
1251831477 M * Snow-Man and the command completes, but if I vserver enter, /mnt is empty..
1251831486 M * Bertl then you bind mount it _again_ in the host namespace
1251831510 M * Bertl what you want to do is to use the _guest_ namespace instead
1251831512 M * Snow-Man doesn't using vnamespace mount it for the guest?
1251831531 M * Bertl and basically you want to mount it in the guest namespace in the first place
1251831561 M * Snow-Man I'd prefer to not restart the vserver, or add capabilities to it, if I don't have to..
1251831563 M * Bertl i.e. no point in mounting it at /a and --bind mounting it then to /b
1251831581 M * Bertl no need to restart or change capabilities for the guest
1251831590 M * Snow-Man ok..  what should I be doing then? :)
1251831619 M * Bertl enter the guest namespace, do the mount into the guest dir (you can use secure-mount to do that for you without messing up the mtab)
1251831676 M * Snow-Man secure-mount?  enter the guest namespace using 'vserver guest enter'?  When I've tried doing mount there, I get permission denied..
1251831711 M * Bertl 'vserver guest enter' enters all spaces and applies all restrictions, i.e. you don't want to do that
1251831728 M * Snow-Man ok..  but not vnamespace either?
1251831731 M * Bertl vnamespace will enter the guest namespace for you, do that, then mount the iso image
1251831796 M * Snow-Man like so?:
1251831797 M * Snow-Man vnamespace -e goldberry mount -o loop /data/vservers/goldberry/home/rootk/debian-502a-amd64-businesscard.iso /data/vservers/goldberry/mnt
1251831821 M * Snow-Man I do that, and then 'vserver goldberry enter', but if I ls -l /mnt from there, it's empty. :/
1251831869 M * Bertl then you must be on debian :)
1251831892 A * Snow-Man smirks.
1251831896 M * ktwilight_ debian is the man!
1251831906 M * Snow-Man yes, I am, but I thought I had fixed the issues around that..
1251831918 M * Snow-Man I'm on 2.6.29.4-vs2.3.0.36.14-beng
1251831930 M * ktwilight_ Bertl, i guess this has to do with vanilla or debian util-vserver?
1251831937 M * Snow-Man and thought I'd fixed up the barrier stuff
1251831938 M * Bertl Snow-Man: util-vserver version?
1251831951 M * Snow-Man 0.30.216~r2772-7
1251831969 M * Bertl probably too old to even give you proper namespace support
1251831979 A * Snow-Man sighs.
1251831987 M * ktwilight_ mine is 0.30.216~r2772-6
1251831991 M * Snow-Man I don't understand why Debian and vservers have some many issues.
1251831993 M * ktwilight_ could it be the kernel patch?
1251831994 M * Bertl but I have no idea what -6 or -7 have
1251832032 M * Snow-Man they're debian-specific and generally don't have changes to code..
1251832035 M * Bertl Snow-Man: because debian is different (and seems to care more about procedures than properly working packages or consistancy or testing :)
1251832073 M * Bertl well, I know that 0.30.26-pre2772 is too old to work with those kernels
1251832080 M * Bertl *0.30.216
1251832106 M * Bertl so if they didn't change anything, I'm not surprised it doesn't work
1251832136 M * ktwilight_ wow, i didn't know there's a 0.40.x
1251832159 M * ktwilight_ uh, nm. i mistaken.
1251832166 M * Snow-Man the latest in Debian is 0.30.216~r2842-2
1251832175 M * Snow-Man would that be more likely to work?
1251832178 M * ktwilight_ pre2841 is out
1251832189 M * ktwilight_ there's no pre2842 :/
1251832279 M * Bertl so if pre2842 works, we will be able to tell when daniel_hozac releases it :)
1251832287 M * Snow-Man The current version of util-vserver is:  0.30.215
1251832290 A * Snow-Man sighs.
1251832329 M * Bertl too old to work with your kernel, or at least not old enough, 0.30.212 will work regarding namespaces (on that kernel) but of course, other stuff will fail
1251832379 M * Snow-Man would 0.30.216~r2842-2 work?
1251832383 M * ktwilight_ do i need to shutdown before upgrading util-vserver? or can i just restart it after upgrading?
1251832403 Q * Piet Ping timeout: 480 seconds
1251832484 M * Bertl Snow-Man: no idea, as pre2842 is not yet released .. but it's probably some version from the repository
1251832500 M * Bertl ktwilight_: restart after upgrading should work
1251832510 M * Snow-Man I imagine it's r2842 from the repo..
1251832623 M * Snow-Man and r2842 didn't change anything.
1251832720 M * Snow-Man other thoughts?
1251832779 M * Bertl try it, let us know
1251832788 M * Snow-Man err, no, I did try it, and it didn't work
1251832788 M * ktwilight_ cool, it works :)
1251832796 M * ktwilight_ Snow-Man, compile from source :)
1251832799 M * Snow-Man at least, using the vnamespace
1251832804 A * Snow-Man sighs.
1251832813 J * BenG ~bengreen@94-169-110-10.cable.ubr22.aztw.blueyonder.co.uk
1251832819 M * Snow-Man ktwilight_: you built pre2841?
1251832821 M * ktwilight_ yup
1251832823 M * ktwilight_ just did it
1251832837 M * Snow-Man what kernel are you running?
1251832892 M * ktwilight_ .27.31
1251832893 M * Bertl ktwilight_: just be careful not to mix the distro version with the self compiled one
1251832910 M * ktwilight_ Bertl, yup, i've first removed the distro version before compiling
1251832928 M * ktwilight_ Snow-Man, vserver is the only package that i don't mind installing from source :)
1251832965 M * Snow-Man based on the changelog, r2842 has only 1 more commit than pre2841, which is "Fix the trailing slash problem."
1251833006 M * Bertl should help not hurt :)
1251833013 M * Snow-Man indeed.
1251833041 J * Piet ~piet@659AAB781.tor-irc.dnsbl.oftc.net
1251833049 M * Snow-Man the debian changelog shows alot of local patches being removed because they're now included upstream as well.
1251833150 A * Snow-Man suspects the problem is somewhere else.
1251833355 M * Bertl the problem being?
1251833370 M * Snow-Man not seeing the files under the mountpoint..?
1251833403 M * Bertl well, if they happen in the correct mount namespace (which if initialized properly, propagates to the guest)
1251833414 J * thierryp ~thierry@home.parmentelat.net
1251833424 M * Bertl then they will show up in the guest as expected
1251833498 M * Snow-Man can I test that vnamespace is working somehow?
1251833513 M * Snow-Man things like vps work fine
1251833564 M * Bertl vps doesn't even touch the mnt namespaces :)
1251833602 M * Bertl and no, we do not have specific test scripts for e.g. vnamespace or namespace isolation yet
1251833664 M * Bertl if you can set the host's hostname from inside a guest, you know that the isolation is broken
1251833695 M * Bertl if you cannot mount into the guest mnt namespace with vnamespace, you know that the mnt namespace setup is broken 
1251833759 M * Bertl note: older utils would work, but mainline decided to change the semantics, so older util-vserver do the wrong thing in recent kernels
1251833776 M * Bertl (would work if it just was for the Linux-VServer API, that is)
1251833788 M * Snow-Man running 'hostname blah' inside the vserver doesn't appear to have any effect on what 'hostname' returns on the host
1251833811 M * Bertl that's a good sign, i.e. you have at least isolation up and running
1251833829 M * Snow-Man using vnamespace or secure-mount doesn't appear to work
1251833846 M * Snow-Man in that, once mounted, I can't see the files from inside the vserver
1251833853 M * Bertl which suggests that the guest namespace setup is broken
1251833899 M * Snow-Man for mounts..  but not for processes?
1251833921 M * Bertl correct
1251833923 M * Snow-Man since vps shows the correct processes (or at least, they match what I see from vserver enter)
1251833951 M * Bertl the process isolation is completely different from the mnt/namespace isolation
1251833975 M * Snow-Man alright, so, what should I be doing to check the mnt/namespace isolation?
1251833989 Q * thierryp Remote host closed the connection
1251834003 M * Snow-Man could this be a chattr issue or something?
1251834068 M * Bertl nope, create a guest (or do the equivalent commands you see in vserver - start --debug), and check that entering the mnt namespace (via vnamespace) and chaning something propagates to the guest ... that's almost the only way to check that it works as expected :)
1251834086 M * Bertl *changing
1251834216 M * Snow-Man heh.
1251834532 M * ktwilight_ Bertl, does testme.sh check on that?
1251834705 M * Bertl nope
1251835361 M * ktwilight_ updated util-vserver by source in the debian section of the wiki :)
1251835375 M * Bertl excellent! tx!
1251836851 Q * BenG Quit: I Leave
1251837318 M * arekm Bertl: was recent v4, v6 binding issue resolved?
1251837869 Q * dna__ Quit: Verlassend
1251837981 M * Bertl arekm: not yet, but I know what causes it
1251838011 M * Bertl but I need to complete a network test script first, otherwise we'll break more than we repair
1251838024 J * geb ~geb@earth.gebura.eu.org
1251838069 Q * bonbons Quit: Leaving
1251838071 J * thierryp ~thierry@home.parmentelat.net
1251838296 M * arekm Bertl: oh, moving back to 27 then
1251838304 M * arekm and waiting 8)
1251838522 Q * hijacker Remote host closed the connection
1251839043 Q * thierryp Remote host closed the connection
1251839331 J * thierryp ~thierry@home.parmentelat.net
1251841113 J * derjohn_mob ~aj@c207002.adsl.hansenet.de
1251841265 N * DoberMann DoberMann[ZZZzzz]
1251841777 Q * thierryp Remote host closed the connection
1251842044 Q * geb Ping timeout: 480 seconds
1251842761 J * thierryp ~thierry@home.parmentelat.net
1251842764 Q * Piet Remote host closed the connection
1251842815 Q * bakins Quit: bakins
1251844838 M * Bertl off to bed now ... have a good one everyone!
1251844846 N * Bertl Bertl_zZ
1251845314 Q * larsivi Ping timeout: 480 seconds