1341551751 Q * clopez Ping timeout: 480 seconds
1341551824 M * Bertl off to bed now ... have a good one everyone
1341551829 N * Bertl Bertl_zZ
1341552891 J * FireEgl ~FireEgl@173-16-9-169.client.mchsi.com
1341553018 M * yang Hi, how can I fix CGROUPS issue http://pastebin.com/F7KhCvJh
1341554231 J * ghislain ~AQUEOS@adsl1.aqueos.com
1341554321 J * ncopa ~ncopa@3.203.202.84.customer.cdi.no
1341558575 Q * cuba33ci Ping timeout: 480 seconds
1341562479 Q * ghislain Quit: Leaving.
1341562634 J * derjohn_mob ~aj@ip-37-201-92-180.unitymediagroup.de
1341564257 M * renihs yang, check if the kernel has this enabled:
1341564258 M * renihs CONFIG_CGROUP_MEM_RES_CTLR=y
1341564258 M * renihs CONFIG_CGROUP_MEM_RES_CTLR_SWAP=y 
1341564706 M * yang CONFIG_CGROUP_MEM_RES_CTLR=y
1341564706 M * yang CONFIG_CGROUP_MEM_RES_CTLR_SWAP=y
1341564706 M * yang # CONFIG_CGROUP_MEM_RES_CTLR_SWAP_ENABLED is not set
1341564706 M * yang # CONFIG_CGROUP_MEM_RES_CTLR_KMEM is not set
1341564881 M * yang renihs: do you think my util-vserver dont have the right flags ?
1341564910 M * renihs hmm make sure the /etc/vserver/<yourguest> name actually matches the guest 
1341564913 M * renihs other than that hmm
1341564937 M * renihs maybe cgroups arent mounted with -o memory?
1341564958 M * renihs else i am out of ideas, someone here certainly knows though :)
1341565889 M * yang maybe cgroups arent mounted with -o memory? where do I perform that ?
1341565911 M * yang it only happens after kernel upgrade
1341566817 J * kir ~kir@swsoft-msk-nat.sw.ru
1341566907 P * kir 
1341566998 M * renihs yang, hmm you could try remounteing it with that 
1341567065 M * renihs but y, dotn think thats even related
1341567072 M * renihs better wait for someone knowledgable :p
1341567117 M * yang ok
1341567522 M * renihs Bertl should wake up anytime soon for abit :)
1341567582 M * yang ;)
1341567679 J * BenG ~bengreen@cpc10-aztw24-2-0-cust114.aztw.cable.virginmedia.com
1341568540 J * dna ~dna@230-171-103-86.dynamic.dsl.tng.de
1341571021 N * Bertl_zZ Bertl
1341571028 M * Bertl morning folks!
1341571121 M * Bertl yang: what kernel/patch/util-vserver version?
1341571384 M * yang morning Bertl 
1341571428 M * Bertl okay so 3.4.1-3.3.3.4 and a recent util-vserver :)
1341571455 M * yang yes
1341571466 M * Bertl is there a special config regarding cgroups on your host?
1341571482 M * Bertl i.e. did you configure certain groups in /etc/vservers or so?
1341571499 M * yang yes i prolly have
1341571504 M * yang i applied the limits
1341571645 M * yang it looks like not on that particular server
1341571668 M * yang I mean I remember defining some "rss" values, but that was prolly another server
1341571717 M * Bertl what does /proc/mounts say on the host regarding cgroups?
1341571779 M * yang on host ?
1341571783 M * yang ok
1341571811 M * yang   /proc/mounts -> self/mounts
1341571831 M * Bertl yeah, cat it and check the content
1341571928 M * yang http://pastebin.com/WfD7Dwx4
1341572003 M * Bertl doesn't look like your cgroups subsystem is mounted
1341572020 M * Bertl this is usually done by the util-vserver runlevel script
1341572030 M * yang oh
1341572036 M * Bertl please double check that it gets executed properly
1341572071 M * yang I cant remember applying special cgroups flags at util-vserver compilation
1341572125 M * Bertl the default should be fine, it will enable all available cgroups
1341572144 M * yang so how can i check if it is compiled with that parameter or if it gets executed
1341572168 M * Bertl for a start, you can run the runlevel script right now
1341572185 M * Bertl it should mount the cgroup subsystem which then should show up in /proc/mounts
1341572348 M * yang soi should basically do what this says http://linux-vserver.org/util-vserver:Cgroups#Ben.27s_install_on_Debian_Squeeze.2FSid
1341572425 M * Bertl if you are on debian, I presume yes
1341572441 M * yang I didnt have /etc/vservers/.defaults/cgroup, nor /lib/udev/devices/cgroup
1341572655 M * yang I guess after applying "vserver    /dev/cgroup    cgroup   cpuset,cpu,cpuacct,devices,freezer,net_cls    0 0" I must reboot now
1341572673 M * yang Do I try something else before reboot ?
1341572761 Q * BenG Quit: I Leave
1341573079 M * yang so actually
1341573080 M * yang # mkdir /dev/cgroup
1341573081 M * yang # mount -t cgroup -o <subsystems> /dev/cgroup
1341573129 M * yang Recent versions of util-vserver sort this out for you by including the appropriate mount command in the util-vserver init (ie: runlevel) script included in the util-vserver distribution, however this apparently only works for the sysv init script, and not the Debian or Gentoo ones.
1341573136 M * yang well I have debian so its different
1341573252 M * Bertl well, I have no idea what debian does differently, but at least on mageia (which uses systemd by default) the util-vserver runlevel scripts get executed as intended (despite the fact that they are sysv ones) 
1341573355 M * Bertl anyway, follow the debian instructions and try again :)
1341573558 M * yang http://pastebin.com/SGVccaJD
1341573644 M * yang I guess i ll just reboot and see after applying debian-specific settings
1341573646 M * Bertl you are missing the 'device'
1341573660 M * Bertl mount -t cgroup -o cpuset,memory none /dev/cgroup
1341573735 M * yang ok i guess i ll add that line to the wiki aswell
1341573788 M * yang i did all that but the error is the same 
1341573810 M * Bertl does /proc/mounts show the cgroup mount now?
1341573812 M * yang But I guess I must implement shares now, so it wont spit out errors
1341573845 M * yang yes
1341573847 M * yang none /dev/cgroup cgroup rw,relatime,memory,cpuset 0 0
1341573926 M * Bertl what does /dev/cgroup contain?
1341573988 M * yang http://pastebin.com/Gm4zWXBs
1341574009 M * Bertl and vserver-stat still says that there is no memory.stat?
1341574020 M * yang yes
1341574029 M * Bertl you did restart the guests, yes?
1341574036 M * yang nope
1341574056 M * Bertl then that's what you want to do, so that they actually use the cgroups
1341574067 M * yang i guess i can try http://linux-vserver.org/util-vserver:Cgroups#Sharing_out_the_CPU_between_guest_servers, then restarting guests and see if it adds limits
1341574099 M * yang I can aswell reboot, if i am already restarting and see if it synces on itself
1341574129 M * Bertl probably won't hurt to check if everything is setup properly in case of system restart
1341574149 M * yang does the cpu shares sum always need to match 100% from all nodes, can I also leave some guests without the memorya limits ?
1341574185 M * Bertl yes, you can leave out the limits for any guest
1341574190 M * yang ok
1341574197 M * Bertl there is a difference between accounting and limits
1341574489 M * yang ok great now it shows perfect
1341574527 M * yang so shares must always match 100% ?
1341574574 M * yang total cpu.shares value being 5120
1341574576 M * Bertl IIRC, they are always hanled in a way that the sum over all is 100%
1341574697 M * yang ok great
1341574727 Q * mharris Ping timeout: 480 seconds
1341574746 M * Bertl but the cgroup system is documented in the kernel's Documentation folder and on the LXC websites
1341574762 M * Bertl Linux-VServer does not modify it from the default behaviour
1341574813 M * yang oh well total sum doesnt need to be 5120, that is only the case for 512 x 10%
1341574835 M * yang It can prolly also be 2048 if i divide 204.8 among 10 VPS's ?
1341574940 M * yang i mean it prolly just counts sizes of the entries among the VPS's with the "cpu.shares" values defined and doesnt count the others
1341575027 J * mharris ~mharris@2001:470:89ad:acdc::f00d
1341575033 M * Bertl Documentation/cgroups/* especially Documentation/cgroups/memory.txt :)
1341575290 M * yang VSZ and RSS values are defined however user/sysTIME still at 0
1341575327 J * clopez ~clopez@131.29.165.83.dynamic.mundo-r.com
1341575514 M * yang brb
1341575517 Q * yang Quit: Free yourself with Free Software, join www.FSF.org
1341576047 J * fisted_ ~fisted@xdsl-87-78-83-226.netcologne.de
1341576059 Q * dna Quit: Verlassend
1341576154 J * bergerx ~bergerx@2-228-78-73.ip190.fastwebnet.it
1341576379 Q * fisted Ping timeout: 480 seconds
1341578874 Q * bergerx Ping timeout: 480 seconds
1341579132 J * yang yang@yang.netrep.oftc.net
1341579144 M * yang Bertl: why don't I get user/sysTIME displayed ?
1341579246 M * daniel_hozac do you have the cpu cgroup enabled?
1341579300 M * yang no
1341579334 M * yang sorry, I have cpu.shares enabled yes
1341579453 M * daniel_hozac and cpuacct?
1341579472 M * yang nope just that
1341579496 M * daniel_hozac that'd be why then.
1341579521 M * yang ok
1341579545 M * yang so i need to umount cgroups and mount with cpuaccts flag
1341579605 M * daniel_hozac the util-vserver initscript will mount it as correctly as possible
1341579632 M * daniel_hozac as correctly as your kernel allows...
1341579955 J * yang_ ~yang@dns3.prunk.si
1341580191 Q * yang Quit: Free yourself with Free Software, join www.FSF.org
1341580255 Q * mcp Quit: ZNC - http://znc.sourceforge.net
1341580265 J * kir ~kir@swsoft-msk-nat.sw.ru
1341580445 P * kir 
1341581569 J * BenG ~bengreen@cpc10-aztw24-2-0-cust114.aztw.cable.virginmedia.com
1341581739 J * mcp ~mcp@wolk-project.de
1341581845 M * yang_ humm
1341581851 M * yang_ how come I keep getting
1341581852 M * yang_ The following problem(s) were encountered while verifying vshelper
1341581852 M * yang_ functionality:
1341581854 M * yang_ * The vshelper state-directory '/var/run/vshelper' does not exist;
1341581869 M * yang_ I didnt get this error earlier when trying to start guests
1341581891 M * yang_ well there is a fix to that down below, but i wonder if it appears because of the cflags
1341581959 M * yang_ or maybe i executed "vprocunhide" from earlier util-vserver
1341581965 M * yang_ triggering that
1341582134 M * Bertl maybe, in general you should avoid mixing util-vserver versions :)
1341582270 M * yang_ i wipe the old one out by executing "make clean" or ?
1341582300 M * daniel_hozac make uninstall
1341582301 M * Bertl if they were installed via 'make install' and you are in the config/build dir of the old install, yes
1341582333 M * Bertl i.e. the new build tree will not uninstall a version properly which had a different config
1341582365 M * yang_ i still have the util-vserver-0.30.216-pre2883 on drive and i removed "util-vserver" debian package previously
1341582515 M * yang_ its a pity that vserver patches got removed from debian
1341582549 M * Bertl well, I presume that debian has a 'better' solution :)
1341582614 J * kir ~kir@swsoft-msk-nat.sw.ru
1341582704 P * kir 
1341582883 Q * BenG Quit: I Leave
1341582945 M * Bertl but honestly, I'm not sure what you are missing ... the broken debian kernels? the evil debian-vserver tools?
1341582976 M * Bertl (or maybe the totally outdated utilities with wrong version numbers?)
1341584055 M * yang_ well i am missing the automatic vserver kernel updates, it used to be simple
1341584067 M * yang_ but anyway got used to compiling out of debian now
1341584071 M * yang_ /lib/util-vserver/defaults/vprocunhide-files
1341584072 M * yang_ /lib/util-vserver/vprocunhide
1341584081 M * yang_ these are the ones that i have outside $HOME
1341584165 Q * renihs Quit: narf
1341584184 M * yang_ there are also some old ones inside /usr/src/util-vserver-0.30.216-pre2883/
1341584227 M * yang_ so can i check on these two if they belong to 3034 branch ?
1341584426 M * yang_ and whats the difference between vprocunhide and vprocunhidefiles
1341585094 M * yang_ it seems that it has gone bad
1341585729 Q * yang_ Quit: leaving
1341585760 J * yang ~yang@199.189.205.50
1341585793 N * yang Guest2334
1341586784 N * ensc Guest2337
1341586794 J * ensc ~irc-ensc@p54ADF515.dip.t-dialin.net
1341587180 Q * brambles Remote host closed the connection
1341587204 Q * Guest2337 Ping timeout: 480 seconds
1341587210 J * brambles brambles@79.133.200.49
1341587504 M * Guest2334 So Bertl , the suggested fix doesn't fix my vshelper
1341587896 M * Guest2334 should i reinstall util-vserver ?
1341588352 N * Guest2334 yang3
1341588383 N * yang3 Guest2342
1341588473 M * Bertl it won't hurt to uninstall, then search for leftovers from the previous install and remove them
1341588480 M * Bertl and then do a clean install
1341588484 M * Bertl off for now .. bbl
1341588487 N * Bertl Bertl_oO
1341590501 Q * derjohn_mob Ping timeout: 480 seconds
1341593532 M * Guest2342 ok fixed now
1341593544 M * Guest2342 even without running "vprocunhide" at reboot
1341593562 M * Guest2342 I purged all previous util-vserver instances
1341593569 M * Guest2342 remade a debian package
1341593579 M * Guest2342 so that it can be uninstalled easier next time
1341593634 J * bergerx ~bergerx@2-228-78-73.ip190.fastwebnet.it
1341594249 M * Bertl_oO yeah, good idea! most of the issues with util-vserver arise from mixed versions because obviously folks do not uninstall old software nowadays :)
1341594531 M * Guest2342 yea
1341594581 J * yang yang@yang.netrep.oftc.net
1341594837 Q * Guest2342 Quit: leaving
1341595054 Q * macmaN Ping timeout: 480 seconds
1341597910 J * macmaN ~chezburge@138.167.190.90.dyn.estpak.ee
1341597950 Q * bergerx Ping timeout: 480 seconds
1341598310 Q * sladen Remote host closed the connection
1341601193 J * dna ~dna@230-171-103-86.dynamic.dsl.tng.de
1341601202 Q * dna 
1341601506 J * hparker ~hparker@2001:470:1f0f:32c:beae:c5ff:fe01:b647
1341601643 M * WMP Bertl_oO: hello, are you alive?
1341602038 M * WMP how to run command on all running vservera?
1341602744 M * yang WMP: which command
1341602758 M * WMP apt-get update
1341602758 P * mharris 
1341602765 M * WMP i try to use vsomething
1341602766 M * yang you need to do it separatelly
1341602771 M * WMP but i dont know how to
1341603133 M * jrayhawk for vserver in $(sudo vserver-stat | tail -n +2 | perl -pe 's/.+ (.+)/\1/'); do sudo vserver $vserver exec apt-get update; done
1341603168 M * WMP only?
1341603174 M * WMP i cant use vsomething?
1341603243 M * jrayhawk Beats me.
1341603310 M * WMP ok
1341603658 J * bonbons ~bonbons@2001:960:7ab:0:c016:f5c9:c1a1:7c59
1341603796 J * hijacker_ ~hijacker@cable-84-43-134-121.mnet.bg
1341605119 M * Wonka is the newest debian vserver kernel still 2.6.32?
1341606501 Q * clopez Ping timeout: 480 seconds
1341606817 J * sladen ~paul@starsky.19inch.net
1341607048 M * fback WMP: vsomething vserver --running -- exec aptitude update 
1341607160 M * WMP nice ;)
1341607162 M * WMP thank ;)
1341609076 Q * hijacker_ Quit: Leaving
1341612122 Q * bonbons Quit: Leaving
1341614370 J * bergerx ~bergerx@2-228-78-73.ip190.fastwebnet.it
1341614776 Q * nou Ping timeout: 480 seconds
1341616809 Q * ensc|w Remote host closed the connection
1341616818 J * ensc|w ~ensc@www.sigma-chemnitz.de
1341617544 Q * bergerx Ping timeout: 480 seconds
1341617674 J * nou Chaton@causse.larzac.fr.eu.org