1266106281 M * petzsch victory: Linux vhost1.kunden-server.org 2.6.32.8-vs2.3.0.36.29 #1 SMP Sun Feb 14 00:53:30 CET 2010 x86_64 GNU/Linux :-D
1266106574 Q * ghislain Quit: Leaving.
1266107480 Q * bonbons Quit: Leaving
1266107821 M * petzsch good night
1266107864 Q * petzsch Quit: Leaving.
1266109816 M * digi1 n8
1266109822 Q * digi1 Quit: Verlassend
1266116451 J * derjohn_foo ~aj@e180193014.adsl.alicedsl.de
1266116884 Q * derjohn_mob Ping timeout: 480 seconds
1266118280 J * aj__ ~aj@e180195217.adsl.alicedsl.de
1266118684 Q * derjohn_foo Ping timeout: 480 seconds
1266119879 Q * Bushmills Quit: Leaving.
1266119904 J * Bushmills ~l@verhau.de
1266120210 M * Bertl off to bed now ... have a good one everyone!
1266120214 N * Bertl Bertl_zZ
1266121931 Q * balbir Read error: Connection reset by peer
1266122757 J * balbir ~balbir@122.172.105.145
1266122758 Q * Bushmills Quit: Leaving.
1266124657 N * DoberMann[PullA] DoberMann[ZZZzzz]
1266124905 Q * quasisane Quit: leaving
1266125280 J * Bushmills ~l@verhau.de
1266125473 Q * Bushmills Remote host closed the connection
1266125593 J * Bushmills ~l@verhau.de
1266125632 Q * Bushmills Remote host closed the connection
1266125654 J * Bushmills ~l@verhau.de
1266125692 Q * Bushmills Remote host closed the connection
1266125723 J * Bushmills ~l@verhau.de
1266125735 Q * Bushmills 
1266125969 J * Bushmills ~l@scarydevilmonastery.net
1266129976 Q * balbir Remote host closed the connection
1266134530 J * ghislain ~AQUEOS@adsl2.aqueos.com
1266136877 J * hijacker_ ~hijacker@87-126-142-51.btc-net.bg
1266137182 Q * hijacker_ Quit: Leaving
1266137713 J * Web-sidux421 ~duke@ip-88-153-237-120.unitymediagroup.de
1266138604 Q * ghislain Quit: Leaving.
1266140245 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d
1266142786 Q * pflanze Read error: Operation timed out
1266143614 J * pflanze ~chris__@69-196-152-229.dsl.teksavvy.com
1266143848 J * Net147 Net147@60-242-179-244.static.tpgi.com.au
1266144026 M * Net147 2.6.32.8-vs2.3.0.36.29 requires CONFIG_CGROUPS and CONFIG_CGROUP_MEM_RES_CTLR to compile now?
1266146456 N * Bertl_zZ Bertl
1266146461 M * Bertl morning folks!
1266146472 M * Bertl Net147: yes, but that will change in the next release
1266146520 M * Net147 is CONFIG_CGROUP_MEM_RES_CTLR_SWAP needed too?
1266146527 M * Bertl yep
1266146715 M * Net147 so in short, everything related to cgroup except DEBUG and NS
1266146792 J * petzsch ~markus@dslb-088-075-164-074.pools.arcor-ip.net
1266146865 M * Bertl yep, the memory/swap accounting and limits where switched to that framework
1266146902 M * Bertl I'll make the code config dependant, so in the next version you can compile with and without it
1266146940 M * petzsch are there any changes to util-vserver required to support that new framework?
1266147020 M * Bertl nope, cgroup support is in recent util-vserver, you just need to change the config (which will be worked around by a mapping in future util-vserver version, if I got daniel_hozac right :)
1266147123 M * petzsch what changes to the config are required? is that documented allready?
1266147123 M * Net147 I still have 2 patches for util-vserver lying around. one fixes cleanup for rpm build during error handling, the other adds support for pacman packages (used by Arch Linux).
1266147233 M * Bertl did you send/submit them to daniel_hozac?
1266147239 M * Net147 some time ago...
1266147246 M * Bertl did he comment on them?
1266147250 M * Net147 the pacman patch was probably updated a tiny bit since then
1266147253 M * Net147 he hasn't replied about the rpm patch
1266147270 M * Net147 same for the last pacman patch I sent
1266147277 M * Bertl and it didn't make it into CVS/SVN either?
1266147279 M * Net147 iirc
1266147290 M * Bertl petzsch: yeah, there is a wiki page about cgroups
1266147296 M * Net147 umm... what's the repo address so I can check?
1266147352 M * Net147 yea it's not in SVN
1266147413 M * petzsch ok, i thought about config changes releated to the update between the latest two development releases
1266147435 J * SauLus ~SauLus@c135033.adsl.hansenet.de
1266147547 M * Bertl petzsch: you have to 'move' rss/vm limits to cgroup mem limits to use them
1266147582 M * Bertl the old framework is still in place but not used for memory virtualization
1266147597 M * Bertl if all works out as expected, we'll remove the old framework soon
1266147778 M * Net147 Bertl: should I send daniel_hozac another email @linux-vserver.org? last I sent was about 3.5 months ago.
1266147879 M * Bertl wouldn't hurt (ad email: http://linux-vserver.org/Developers)
1266148182 M * Net147 done
1266149517 Q * Net147 Quit:  HydraIRC -> http://www.hydrairc.com <-
1266150435 Q * theocrit1 Quit: leaving
1266150618 M * petzsch this is weired: i've got virt_mem flag set and rss.hard limited, but in the VPS it shows me the total host memory as free and 0k as used
1266150770 M * Bertl not weird, expected :0
1266150787 M * Bertl as I said, memory virtualization is based on the cgroup limits
1266150804 M * Bertl i.e. forget about rss.hard/soft, configure the cgroup memory limits
1266150839 M * Mr_Smoke IS that thing about CGROUPS_NS needing to be disabled still true ?
1266150840 M * petzsch then i misunderstood you: i thought both where active now and rss.hard would be removed soon
1266150844 M * Mr_Smoke It's on the wiki page for cgroups
1266150880 M * Bertl Mr_Smoke: yes
1266150907 M * Bertl petzsch: rss.hard still works, but it isn't used for the virt_mem anymore
1266150936 M * Mr_Smoke Ok then.
1266150942 M * petzsch so the memory limit is in place, but just not displayed to the user? ok, i'll see what the flower page has in store for me ;-)
1266150956 M * Bertl petzsch: correct
1266152085 M * petzsch so much to the flower page: This directory contains cgroup settings to be applied to this guest. See your kernel documentation for what settings are valid with your configuration. ... i'll dig myself through the kernel docs and hopefully drop some lines in the wiki by this evening
1266152095 M * petzsch gotta go get a train
1266152440 Q * petzsch Quit: Leaving.
1266154700 J * petzsch ~markus@dslb-088-075-164-074.pools.arcor-ip.net
1266155633 J * Piet ~Piet__@04ZAAADPK.tor-irc.dnsbl.oftc.net
1266156667 Q * Mr_Smoke Read error: Connection reset by peer
1266156713 J * Mr_Smoke smokey@layla.lecoyote.org
1266157823 Q * manana Remote host closed the connection
1266157904 J * manana ~mayday090@84.17.11.51
1266159145 Q * petzsch Quit: Leaving.
1266160066 J * hijacker_ ~hijacker@87-126-142-51.btc-net.bg
1266160630 Q * Web-sidux421 Remote host closed the connection
1266161401 J * petzsch ~markus@184.121.113.82.net.de.o2.com
1266161668 M * petzsch positive: cpu limits seem to work more precise now: my 50% limitation is causing %cpu to keep between 50 and 51% *happy*
1266161695 J * dna ~dna@170-198-103-86.dynamic.dsl.tng.de
1266163240 Q * petzsch Quit: Leaving.
1266163458 J * fLoo ~fLoo@fs.coresec.de
1266163905 Q * Bushmills Quit: Leaving.
1266163961 J * Bushmills ~l@78.46.19.202
1266164050 Q * Bushmills 
1266164071 J * Bushmills ~l@78.46.19.202
1266164293 J * pmenier ~pmenier@ACaen-152-1-92-152.w83-115.abo.wanadoo.fr
1266164610 Q * hijacker_ Remote host closed the connection
1266165502 J * petzsch ~markus@82.113.106.188
1266165959 Q * petzsch Quit: Leaving.
1266167618 M * pflanze What's the right way to measure total cpu usage on a system (including kernel activity)?
1266167649 M * pflanze (not %, but total cpu time)
1266167904 M * Bertl time?
1266171008 J * quasisane ~sanep@c-75-67-251-206.hsd1.nh.comcast.net
1266171115 J * theocrite ~Hubert@kim.theocrite.org
1266172028 Q * pmenier Quit: Konversation terminated!
1266173458 J * derjohn_foo ~aj@e180197220.adsl.alicedsl.de
1266173820 Q * aj__ Ping timeout: 480 seconds
1266174694 Q * bonbons Quit: Leaving
1266174831 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d
1266176531 J * ghislain ~AQUEOS@adsl2.aqueos.com
1266176935 M * Bertl off to bed now ... have a good one everyone!
1266176940 N * Bertl Bertl_zZ
1266177271 Q * ghislain Quit: Leaving.
1266179320 J * BenG ~bengreen@cpc2-aztw22-2-0-cust521.aztw.cable.virginmedia.com
1266179665 Q * BenG Quit: I Leave
1266180045 Q * dna Quit: Verlassend
1266185165 Q * bonbons Quit: Leaving
1266185773 M * Bushmills i just notice that recent packaging of munin comes with vserver plugin already, no need to write one.
1266185834 M * Bushmills pflanze: total at any one given moment, or with a recorded history? 
1266185858 M * pflanze total at some moment (like sum since boot or so)
1266185889 M * Bushmills "since" implies history
1266185896 M * Bushmills "now" or "not only now"
1266185928 M * Bushmills top  can show for the last x seconds, according the time interval you have set
1266185932 M * pflanze Just the sum of time a core spent not sleeping, separated into different buckets for the frequencies it is running at.
1266185944 M * pflanze For all cores.
1266185961 M * pflanze Yeh, I should probably look how top does it.
1266185982 M * Bushmills afaik, top shows sum for all cores
1266186035 M * Bushmills per process, ps
1266186053 M * Bushmills (cumulated)
1266186067 M * pflanze I want it for everything, because some heavy activities in the kernel are not accounted to the processes causing it.
1266186113 M * pflanze But actually I've already finished the work I would have needed it for, I just did it by watching top's percentage and use that :/.
1266186122 M * pflanze It still would be useful to know for next time.
1266186857 M * Bushmills pflanze: for graphing cpu load, the abovely mentioned munin is nice
1266186869 M * Bushmills gives histroy in day, week, month and year
1266186912 M * pflanze I'm mostly interested in how to get the value out of the kernel.
1266186947 M * pflanze I could study sources of such programs of course; maybe I'll find out anyway some time.
1266187020 M * Bushmills the munin plugin gets it from /proc/stat
1266187079 M * pflanze Ahh. Simple, toplevel proc directory; I was looking for cpu related stuff further below.
1266187115 M * pflanze Thanks for the tip. Reading the kernel sources for that file should make things clear.
1266187183 M * Bushmills yw
1266187832 Q * agaffney Quit: moving irssi to a different box
1266188019 J * gdm ~gdm@lair.fifthhorseman.net
1266188800 J * agaffney ~agaffney@ryoko.agaffney.org
1266188831 Q * agaffney 
1266188836 J * agaffney ~agaffney@ryoko.agaffney.org
1266189708 Q * Piet Quit: Piet
1266189740 J * Piet ~Piet__@04ZAAAD8A.tor-irc.dnsbl.oftc.net
1266189771 J * petzsch ~markus@p4FF45565.dip.t-dialin.net