1332376477 J * BenG ~bengreen@cpc10-aztw24-2-0-cust114.aztw.cable.virginmedia.com 1332377098 J * clopez ~clopez@82.25.60.213.dynamic.mundo-r.com 1332377343 Q * clopez 1332377360 J * clopez ~clopez@82.25.60.213.dynamic.mundo-r.com 1332378301 Q * BenG Quit: I Leave 1332381002 M * Bertl off to bed now ... have a good one everyone! 1332381011 N * Bertl Bertl_zZ 1332389325 Q * DreamerC Ping timeout: 480 seconds 1332389555 J * DreamerC ~DreamerC@122-116-181-118.HINET-IP.hinet.net 1332395289 Q * clopez Ping timeout: 480 seconds 1332396804 J * voegelas ~voegelas@HSI-KBW-109-192-051-187.hsi6.kabel-badenwuerttemberg.de 1332401565 Q * derjohn_mob Ping timeout: 480 seconds 1332401884 J * ghislain ~AQUEOS@adsl2.aqueos.com 1332403364 Q * jeroen__ Remote host closed the connection 1332404226 J * jeroen__ ~jeroen@095-097-051-172.static.chello.nl 1332406426 J * hola c193def5@ircip3.mibbit.com 1332407999 J * spechtacular 182b2f08@ircip1.mibbit.com 1332413121 N * Bertl_zZ Bertl 1332413126 M * Bertl morning folks! 1332416495 J * clopez ~clopez@82.25.60.213.dynamic.mundo-r.com 1332418871 J * derjohn_mob ~aj@87.253.171.221 1332419124 Q * clopez Ping timeout: 480 seconds 1332419976 M * z0d morning 1332419989 Q * PowerKe Ping timeout: 480 seconds 1332423239 M * ncopa hi 1332423255 M * ncopa i wonder if it would be an idea to rename the gentoo initstyle to "openrc" 1332423310 M * ncopa alpine linux uses openrc too 1332423722 J * PowerKe ~tom@94-226-105-186.access.telenet.be 1332425132 J * kir ~kir@swsoft-msk-nat.sw.ru 1332425224 J * clopez ~clopez@155.99.117.91.static.mundo-r.com 1332428035 J * dowdle ~dowdle@scott.coe.montana.edu 1332430077 M * ghislain for compatibility perhaps making it an alias has changing would break some setup :) 1332431035 Q * derjohn_mob Ping timeout: 480 seconds 1332431102 P * kir PING 1332431102 1332431139 Q * dowdle 1332431312 J * dowdle ~dowdle@scott.coe.montana.edu 1332432745 Q * spechtacular Quit: http://www.mibbit.com ajax IRC Client 1332434525 J * fleischergesell ~fleischer@p5B0A1581.dip.t-dialin.net 1332434704 M * fleischergesell Hey, is the flag VIRT_CPU broken in recent builds? I notice a log of noise when watching htop that does not origin from the guest I entered. Also if I set cpuset.cpus to"6-7" I can still see all 8 cores in the guest. Is this expected bevahior? 1332435072 Q * voegelas Quit: Leaving. 1332435215 M * Bertl VXF_VIRT_CPU was removed around 2.6.23 1332435238 M * Bertl cpu limits and isolation was replaced with cgroups 1332435244 M * fleischergesell I use cgroups 1332435304 M * fleischergesell All I want is to not see the cpus that are not in the cpuset I assigned to a g uest 1332435312 M * fleischergesell Is that possible? 1332435316 M * Bertl yes 1332435330 M * fleischergesell Would you provide me with a link or howto? 1332435393 M * fleischergesell Okay, 'all' was not entirely true - I also do not want any guest to see cpu usage _not_ origniation from itself 1332435409 M * Bertl http://vserver.13thfloor.at/ExperimentalT/delta-cpuset-feat01.diff 1332435435 Q * Alex[fob] Ping timeout: 480 seconds 1332435501 M * fleischergesell Okay, I'm no C programmer but should I add something like cpuset.mask to the guest cgroups or similar? 1332435548 M * Bertl if you want to limit the guest to certain cpus, yes 1332435584 M * fleischergesell Is there something like a list of what cgroups I can use vor guests? 1332435596 M * fleischergesell *for 1332435607 M * Bertl basically all of them are supported with recent util-vserver 1332435619 M * Bertl see the 'Great Flower Page' for config details 1332436074 M * Bertl off for a nap ... bbl 1332436080 N * Bertl Bertl_zZ 1332436104 M * fleischergesell Thanks for the hint, but I still cannot find the appropriate cgroup to virtualize cpu usage in a guest - and I dont find the cgroup to make all cpus that are not in the current set hidden 1332436613 Q * ensc|w Remote host closed the connection 1332436621 J * ensc|w ~ensc@www.sigma-chemnitz.de 1332436621 J * bonbons ~bonbons@2001:960:7ab:0:5da3:7332:299e:412c 1332438565 Q * hola Quit: http://www.mibbit.com ajax IRC Client 1332443848 N * Bertl_zZ Bertl 1332443854 M * Bertl back now ... 1332443988 J * derjohn_mob ~aj@80.187.201.77 1332445324 Q * theocrite Ping timeout: 480 seconds 1332445369 J * theocrite ~Hubert@kim.theocrite.org 1332445521 J * NightMonkey ~NightMonk@nightmonkey.user.oftc.net 1332445554 M * NightMonkey Howdy, are there any "sneakier" ways to kill processes in vservers that don't die, other than vserver stop and vkill -s 9 ? 1332445975 M * Bertl vkill without stopping the guest? 1332446010 M * NightMonkey Sadly, this is the last process left standing after vserver stop, so that's why I'm grasping at straws. 1332446047 M * Bertl so, what process is it? 1332446070 M * NightMonkey 19453 41 devadmin01 ? R 0:00 /usr/sbin/httpd.worker 1332446150 M * Bertl interesting, why doesn't it die/shut down when the guest is stopped? 1332446160 M * NightMonkey Heh, I wish I knew. 1332446193 M * Bertl well, what init style are you using? sysv? 1332446203 M * NightMonkey I've had this problem for a while, but I did upgrade vserver. 1332446205 M * NightMonkey Ah, let me check. 1332446335 M * NightMonkey Hrm, it's not explicitly set. 1332446342 M * Bertl then it should be sysv 1332446346 M * NightMonkey "default" 1332446392 M * Bertl so, in this case, the sysv runlevel script (for httpd?) should stop the processes 1332446399 M * NightMonkey Yeah. 1332446415 M * Bertl I presume it writes something like: stopping httpd .... 1332446425 M * NightMonkey Something has gone pear-shaped, though, because our other vservers do stop. 1332446445 M * NightMonkey Oh, yes, I see the init runlevel change output. 1332446465 M * Bertl so, any errors reported there? and what are the involved kernel/patch versions? 1332446575 M * NightMonkey Kernel - 2.6.32-220.7.1.el5.vs2.3.0.36.29.6.22.x86_64 1332446589 M * Bertl that's the old one, yes? 1332446595 M * NightMonkey The errors are that a timeout occoured and it couldn't kill the last process. 1332446632 M * NightMonkey Odd, we're using dhozac-vserver repository for yum. 1332446651 M * NightMonkey But, I do see from the /topic that we're apparently behind? 1332446696 M * NightMonkey Oh, wait, we're actually ahead, based on /topic (stable). 1332446748 M * NightMonkey OK, sadly, I've got developers eating their livers because this vserver is our authentication instance, so I have to punt and reboot the host. 1332446857 M * Bertl what's the start/stop output you get for this guest? 1332446881 M * NightMonkey One moment, I'll gather. 1332446972 M * NightMonkey Ah, sorry, it has scrolled out of my terminal buffer. :/ 1332446987 M * Bertl get a better terminal then :) 1332447018 M * NightMonkey Yeah, my screen setup needs another zero on the end of its scrollback definition... 1332447038 J * hijacker_ ~hijacker@cable-84-43-134-121.mnet.bg 1332447046 M * NightMonkey Anyway, thanks for being a soundingboard. 1332447084 M * Bertl np, I'd look for errors on shutdown or missing services 1332447093 M * NightMonkey But, there's no way, other than the vserver tools, to "surgically" remove a process for a vserver that's "stopped"? 1332447100 M * Bertl i.e. services which are reported starting on guest startup 1332447117 M * NightMonkey It's just odd that it is in a "Running" state, and doesn't care about vkill -s -9. 1332447124 M * Bertl well, the guest isn't stopped as long as there is still a process left 1332447146 M * NightMonkey Is there a way to simulate a "pull the power cord" event? ;) 1332447155 M * Bertl that's what vkill is for 1332447165 M * NightMonkey OK, that's what I was afraid of. :) 1332447168 M * Bertl but it won't work on processes stuck in 'D' state 1332447188 M * Bertl (which usually mean that the kernel has/had some problems) 1332447331 M * NightMonkey Odd. 1332447370 M * NightMonkey I'd be less worried, except that this process was reported to be in the "R"unning state. Oh, well, I stomped on the evidence. 1332447398 M * Bertl running should be fine and vkill will work on those 1332447526 P * fleischergesell 1332447648 M * NightMonkey Unfortunately, it didn't in this case. 1332447793 J * Aiken ~Aiken@2001:44b8:2168:1000:21f:d0ff:fed6:d63f 1332450355 Q * clopez Ping timeout: 480 seconds 1332451038 Q * ghislain Quit: Leaving. 1332452221 Q * bonbons Quit: Leaving 1332452533 Q * hijacker_ Quit: Leaving 1332453703 J * morfoh ~morfoh@shell.opensde.net 1332453717 M * morfoh Hi 1332453922 M * morfoh Bertl: http://nopaste.opensde.net/?YmJlMT <-- there you will find infos of what have to be adjusted within patch-3.0.24-vs2.3.2.3.diff to cleanly apply on 3.0.25 1332453954 M * morfoh this time no diff of the diff ;) 1332454669 Q * mnemoc Remote host closed the connection 1332454669 Q * karasz Remote host closed the connection 1332454811 M * morfoh oha... seems I scared my friends ;) 1332454832 M * morfoh so ... time to go sleep too 1332455331 J * mnemoc ~amery@shell.opensde.net 1332455565 M * Bertl morfoh: tx 1332455918 Q * derjohn_mob Ping timeout: 480 seconds 1332455968 Q * mnemoc Quit: leaving 1332456297 J * karasz ~karasz@shell.opensde.net 1332457186 Q * dowdle