1337128666 Q * ichavero_ Quit: This computer has gone to sleep 1337130383 Q * clopez Ping timeout: 480 seconds 1337130781 J * ichavero_ ~ichavero@189.231.61.252 1337130796 Q * ichavero_ 1337134247 J * zzach1 ~zzach@dslb-178-006-086-228.pools.arcor-ip.net 1337134602 Q * zzach Ping timeout: 480 seconds 1337146024 M * Bertl off to bed now ... have a good one everyone! 1337146030 N * Bertl Bertl_zZ 1337148463 J * ncopa ~ncopa@3.203.202.84.customer.cdi.no 1337148471 J * ghislain ~AQUEOS@adsl2.aqueos.com 1337148967 Q * derjohn_mob Ping timeout: 480 seconds 1337150579 Q * ncopa Ping timeout: 480 seconds 1337150868 J * ncopa ~ncopa@ti0143a340-0424.bb.online.no 1337152373 Q * FireEgl Remote host closed the connection 1337153276 J * FireEgl ~FireEgl@173-16-9-169.client.mchsi.com 1337154555 Q * FireEgl Remote host closed the connection 1337155460 J * FireEgl ~FireEgl@173-16-9-169.client.mchsi.com 1337157863 J * uranus ~uranus@62.152.161.117 1337157865 M * uranus Hi all 1337157939 M * uranus i have a strange issue with kernel 3.3.6 and uv pre3034: if i reboot a guest with reboot -f, all processes which were started through init does belong to the root cgroup and not to the guest cgroup 1337158255 J * arthus ~arthus@reverse-232.fdn.fr 1337160165 J * kir ~kir@swsoft-msk-nat.sw.ru 1337160481 J * ghislain1 ~AQUEOS@adsl1.aqueos.com 1337160481 Q * ghislain Read error: Connection reset by peer 1337160677 Q * FireEgl Ping timeout: 480 seconds 1337162289 J * BenG ~bengreen@cpc10-aztw24-2-0-cust114.aztw.cable.virginmedia.com 1337162727 J * clopez ~clopez@fanzine.igalia.com 1337163623 Q * brambles Remote host closed the connection 1337165214 J * benl ~benl@dockoffice.sonassihosting.com 1337165650 J * brambles brambles@79.133.200.49 1337165719 Q * nkukard Quit: Leaving 1337166862 J * fisted_ ~fisted@xdsl-84-44-144-190.netcologne.de 1337167017 Q * fisted Ping timeout: 480 seconds 1337167224 Q * ensc|w Remote host closed the connection 1337167232 J * ensc|w ~ensc@www.sigma-chemnitz.de 1337168099 Q * BenG Quit: I Leave 1337168931 N * Bertl_zZ Bertl 1337168948 M * Bertl morning folks! 1337168969 M * Bertl uranus: guest config? 1337169449 Q * jeroen__ Remote host closed the connection 1337170118 M * benl Morning 1337170166 M * uranus Bertl, nothing special, style => sysv and memory limit with cgroup, the rest is a verver build skeleton config 1337170600 M * Bertl on a sysv setup, init does not start any guest processes 1337170626 M * Bertl (mainly because there is no guest init) 1337170885 M * uranus well but reboot -f restarts all services 1337170893 M * uranus and ssh sessions got disconnected 1337170942 M * Bertl reboot -f will call into the host (default setup) and util-vserver will do a guest restart 1337170987 M * uranus then there is something broken with uril vserver if all started processes belong to the root cgroup? 1337170991 M * Bertl after that guest services should end up in the guest cgroup(s), if not, please file a bug report against util-vserver (or talk to daniel_hozac) 1337171032 M * daniel_hozac Bertl: do we run vshelper in the root cgroup? 1337171064 M * Bertl most likely, it is a kernel helper 1337171099 M * Bertl but let me double check 1337171686 M * Bertl the helper should be executed withing the khelper_wq, which is singlethreaded kernel workqueue 1337171786 M * Bertl if that executes outside the root cgroup, something is wrong I'd say 1337172145 Q * Aiken Remote host closed the connection 1337172200 M * Bertl just verified on a slightly older kernel it executes in the root cgroup 1337172208 M * daniel_hozac yeah. 1337172220 M * Bertl (by adding a logger "`cat /proc/self/cgroup`" to the vshelper 1337172258 M * uranus so this means that util-vserver is working as expected? 1337172465 M * Bertl well, at least this part is working as expected :) 1337172506 M * uranus ok so util vserver might not assign the guest cgroup back because vshelper is executed in the root cgroup? 1337172598 M * ncopa thanks for 3.3.6 patch! 1337172810 M * daniel_hozac uranus: could you try touch /etc/vservers/.defaults/apps/vshelper/debug /etc/vservers/.defaults/apps/vshelper/logfile 1337172834 M * daniel_hozac uranus: and editing /sbin/vshelper to have --debug in the $_VSERVER command on line 61 1337172857 M * uranus just a moment 1337172901 M * uranus --debug is only needed in the restart case? 1337172908 M * daniel_hozac yes. 1337172946 M * daniel_hozac then paste the contents of /etc/vservers/.defaults/apps/vshelper/logfile after a reboot -f from a guest. 1337173126 M * uranus daniel_hozac, http://paste.linux-vserver.org/22745 1337173982 Q * ncopa Quit: Leaving 1337174188 J * ncopa ~ncopa@3.203.202.84.customer.cdi.no 1337174384 M * daniel_hozac uranus: that got cut off. 1337174448 M * uranus oh 1337174537 M * uranus daniel_hozac, part 2: http://paste.linux-vserver.org/22746 1337174590 M * uranus daniel_hozac, part 3: http://paste.linux-vserver.org/22748 1337174660 M * Bertl multipart pastebin :) 1337174699 M * uranus ;) 1337174746 M * uranus the output is to long :) 1337174779 M * Bertl pick out the relevant parts :) 1337174798 M * uranus what are the relevant parts? 1337175122 M * daniel_hozac basically the last 200-300 lines. 1337175164 M * uranus that would be part 3 1337175260 M * daniel_hozac hmm 1337175272 M * daniel_hozac your bash doesn't error on the fact that it failed to write to the file. 1337175344 M * uranus there is bash version 4.1 from debian squeeze installed 1337175823 M * Bertl so the problem is a buggy bash? 1337175866 M * uranus another problem is, why can't the bash write to the tasks file 1337175896 J * vspas ~vspas@87.213.36.165 1337176225 M * uranus this means after all that cgroup does have a problem here? 1337176293 M * daniel_hozac yes 1337176424 M * uranus the strange thing is: if i use vserver restart all is setup correct 1337176468 M * uranus it does only not work when called via reboot -f 1337176602 M * daniel_hozac that's not too unexpected. 1337176652 M * uranus but because of what? 1337176842 M * uranus the bash even doesn't error after touch /etc/vservers/testguest/nonexistend 1337176888 M * daniel_hozac why would it? 1337176921 M * uranus i thought it should error because of your statement: your bash doesn't error on the fact that it failed to write to the file. 1337177438 M * Bertl daniel_hozac: so, what's the problem as far as you can tell? 1337177472 M * daniel_hozac i'm guessing one of the subsystems is rejecting the attachment for some reason. 1337177477 M * daniel_hozac why i have no idea yet. 1337177488 M * daniel_hozac (building a debug kernel to figure out which subsystem now) 1337177515 J * nkukard ~nkukard@197.87.42.242 1337178111 J * ghislain ~AQUEOS@adsl1.aqueos.com 1337178453 Q * ghislain1 Ping timeout: 480 seconds 1337179568 Q * transacid Remote host closed the connection 1337179576 J * dowdle ~dowdle@scott.coe.montana.edu 1337179832 N * ensc Guest305 1337179841 J * ensc ~irc-ensc@p54ADDC48.dip.t-dialin.net 1337179951 J * transacid ~transacid@transacid.de 1337179975 N * transacid Guest306 1337180274 Q * Guest305 Ping timeout: 480 seconds 1337182584 J * ghislain1 ~AQUEOS@adsl1.aqueos.com 1337182884 Q * ghislain Ping timeout: 480 seconds 1337183048 Q * vspas Ping timeout: 480 seconds 1337183836 J * ghislain ~AQUEOS@adsl1.aqueos.com 1337184148 Q * ghislain1 Ping timeout: 480 seconds 1337184254 J * ghislain1 ~AQUEOS@adsl2.aqueos.com 1337184497 Q * ghislain Ping timeout: 480 seconds 1337184910 J * bonbons ~bonbons@2001:960:7ab:0:c53b:b9c:9130:893 1337185456 M * Bertl off for now ... bbl 1337185461 N * Bertl Bertl_oO 1337186092 M * ghislain1 bertl: 3.2.17 compile and run fine, tested on a r300 everything fine until now 1337189112 M * uranus is there any chance to define one ipv6 ip per guest as ip to use for outgoing connections? 1337189597 Q * ncopa Quit: Leaving 1337191224 J * fisted ~fisted@xdsl-87-78-184-98.netcologne.de 1337191239 Q * fisted_ Ping timeout: 480 seconds 1337191631 M * daniel_hozac uranus: i'm not sure i understand your question. 1337191832 M * uranus if I assign multiple ipv6 addresses to a guest, it is more or less a gamble which ipv6 address is used for outgoing connections 1337191893 M * daniel_hozac it's the last one added in Linux. 1337192099 M * uranus ah thx 1337192328 Q * benl Quit: HydraIRC -> http://www.hydrairc.com <- 1337193072 Q * fisted Quit: leaving 1337193077 J * fisted ~fisted@xdsl-87-78-184-98.netcologne.de 1337193553 Q * clopez Ping timeout: 480 seconds 1337196392 Q * imcsk8 Ping timeout: 480 seconds 1337196980 J * imcsk8 ~ichavero@148.229.1.11 1337197499 J * hijacker_ ~hijacker@cable-84-43-134-121.mnet.bg 1337199947 Q * hijacker_ Quit: Leaving 1337200152 Q * dowdle 1337201845 M * daniel_hozac Bertl_oO: the vshelper problem is because it is launched the way it is. 1337201871 M * daniel_hozac it gets a PF_THREAD_BOUND in task->flags, which makes the cpuset cgroup reject the attach. 1337202042 M * daniel_hozac i'm not entirely sure how we should fix that... 1337203787 Q * bonbons Quit: Leaving 1337203884 J * Aiken ~Aiken@2001:44b8:2168:1000:21f:d0ff:fed6:d63f 1337203904 J * clopez ~clopez@200.19.165.83.dynamic.mundo-r.com 1337207624 Q * fisted Ping timeout: 480 seconds 1337208272 Q * arthus Quit: Lost terminal