1269734993 J * Piet_ ~Piet__@04ZAABCHT.tor-irc.dnsbl.oftc.net 1269735380 Q * Piet Ping timeout: 480 seconds 1269737442 Q * ghislain Quit: Leaving. 1269738044 Q * SlackLnx Quit: I'll Be Back 1269738170 J * ichavero_ ~ichavero@8.14.254.228 1269738779 N * DoberMann[PullA] DoberMann[ZZZzzz] 1269740003 Q * bobnormal Ping timeout: 480 seconds 1269740006 J * bobnormal ~irc@c122-106-251-71.belrs3.nsw.optusnet.com.au 1269741932 Q * ichavero_ Ping timeout: 480 seconds 1269742940 Q * Piet_ Ping timeout: 480 seconds 1269743509 J * Piet_ ~Piet__@04ZAABCJQ.tor-irc.dnsbl.oftc.net 1269743901 J * skibo ~skibo@cpe-68-206-112-40.stx.res.rr.com 1269744036 M * skibo anybody alive 1269744059 M * Bertl yup, what's up? 1269744088 M * skibo nothing much just wanted to see who is all awake...LOL 1269744109 M * skibo where u from? 1269744228 M * Bertl from a place where it is considered impolite to bother folks because you are bored :) 1269744283 M * skibo oh not bored... just trying to see what is everybody up too just making small talk.. 1269744299 M * skibo but it seems like your pretty much being bothered.... 1269744321 M * skibo by it...so i'll leave you alone... 1269744369 M * Bertl k, appreciated ... let me know if you have any Linux-VServer related questions or topics ... 1269744424 M * skibo actually i do 1269744461 M * skibo i just dont come out to bother ppl... 1269744487 M * Bertl let's hear then 1269744515 M * skibo never mind let me help this other person with a simple question on another channel 1269744521 P * skibo Leaving 1269744644 M * Bertl why am I not surprised :) 1269745232 J * aj__ ~aj@e180194045.adsl.alicedsl.de 1269745251 J * SauLus_ ~SauLus@c150174.adsl.hansenet.de 1269745660 Q * SauLus Ping timeout: 480 seconds 1269745660 N * SauLus_ SauLus 1269745665 Q * derjohn_foo Ping timeout: 480 seconds 1269751073 Q * nkukard Ping timeout: 480 seconds 1269756404 M * Bertl off to bed now ... have a good one everyone! 1269756413 N * Bertl Bertl_zZ 1269761876 J * nkukard ~nkukard@196-210-44-84.dynamic.isadsl.co.za 1269763153 J * ntrs ~ntrs@77.29.202.151 1269764155 J * ntrs_ ~ntrs@77.29.203.184 1269764587 Q * ntrs Ping timeout: 480 seconds 1269764931 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d 1269765588 Q * bonbons Quit: Leaving 1269765839 J * balbir ~balbir@122.172.166.81 1269767238 J * hijacker_ ~hijacker@87-126-142-51.btc-net.bg 1269767401 J * ghislain ~AQUEOS@adsl2.aqueos.com 1269770732 J * petzsch ~markus@p4FF4730C.dip.t-dialin.net 1269772040 J * sleblanc ~serge@LAubervilliers-153-53-1-14.w217-128.abo.wanadoo.fr 1269772307 M * sleblanc Est-ce que l'un d'entre vous est francophone ou esperantophone ? 1269772317 Q * hijacker_ Quit: Leaving 1269772381 M * sleblanc Does anyome would like help m 1269772427 M * sleblanc i about ubuntu upstart udev and rsyslog daemon ? 1269772672 Q * ntrs_ Ping timeout: 480 seconds 1269772749 M * bobnormal sleblanc: http://linux-vserver.org/Upstart_issues 1269773662 Q * derjohn Quit: by(t)e 1269774002 M * sleblanc Mi faris la ordonojn skribitajn en la dosiero Upstart_issues. Sed poste, la programo aptitude fiksiĝas kun la eraro : Errors were encountered while processing: 1269774002 M * sleblanc udev 1269774002 M * sleblanc rsyslog 1269774243 M * sleblanc Ĉu mi devas formeti la progamojn udev kaj rsyslog ? 1269775514 M * biz sleblanc: udev errors are probably related to a failing upstart (udev is started via an upstart event). rsyslog will fail in a guest trying to access /proc/kmsg, just disable the imklog module in rsyslog.conf 1269777558 M * sleblanc Upstart seems good work, but the wrong appears when i use Aptitude. 1269777618 Q * balbir Read error: Connection reset by peer 1269777732 M * sleblanc Aptitude freeze in post-installation of udev and rsyslog. 1269778002 M * sleblanc Aptitude freeze during an post-installation of an udev and rsyslog update. 1269778416 J * balbir ~balbir@122.172.164.67 1269779143 Q * balbir Ping timeout: 480 seconds 1269779673 J * balbir ~balbir@122.167.251.79 1269781647 Q * balbir Read error: Connection reset by peer 1269782065 J * ntrs ~ntrs@77.29.15.145 1269782519 J * balbir ~balbir@122.166.160.121 1269782708 Q * manana Remote host closed the connection 1269782827 J * manana ~mayday090@84.17.25.149 1269784453 N * Bertl_zZ Bertl 1269784458 M * Bertl morning folks! 1269784509 M * petzsch moin Bertl 1269784833 M * Bertl sleblanc: you are updating an ubuntu guest or what? 1269785566 Q * Chlorek Ping timeout: 480 seconds 1269785717 J * Chlorek ~cokolwiek@c.sed.pl 1269785835 Q * aj__ Ping timeout: 480 seconds 1269786202 Q * Chlorek Quit: - 1269786210 J * Chlorek ~cokolwiek@c.sed.pl 1269786518 Q * Chlorek Quit: - 1269786521 J * Chlorek ~cokolwiek@c.sed.pl 1269786876 J * dna ~dna@31-216-103-86.dynamic.dsl.tng.de 1269787098 Q * balbir Read error: Connection reset by peer 1269787563 M * petzsch i've got a question concerning http://linux-vserver.org/util-vserver:Cgroups#Draft_-_Distributing_cpu_shares_with_cgroups : is there a reason why the example uses 2^n values or could i just use "1" and "1" for two vservers with 50% CPU share each? 1269787628 M * Bertl the only reason probably is that the default is 1024 or so 1269787929 J * balbir ~balbir@122.167.171.210 1269787959 Q * ktwilight Read error: Connection reset by peer 1269788116 J * ktwilight ~keliew@91.179.218.231 1269788410 M * petzsch working like a charme 1269789068 J * hijacker_ ~hijacker@87-126-142-51.btc-net.bg 1269789117 J * derjohn_mob aj@80.187.212.13 1269789877 Q * derjohn_mob Ping timeout: 480 seconds 1269790023 J * oli ~oli@62.152.161.117 1269790047 Q * petzsch Quit: Leaving. 1269790054 M * oli hi all, is there a possibility to kill a context which does not stop with vserver stop 1269790092 M * oli the context has many time the same processes running: reboot -d -f -i 1269790110 M * oli Vserver still running unexpectedly; please investigate it manually... 1269790185 M * Bertl you can kill the processes with vkill 1269790202 M * oli that doesn't help anything the process is still there after killing 1269790205 M * Bertl once the last process exits, the context will go away too 1269790237 M * Bertl what state are the processes in? 1269790253 M * oli D 1269790302 M * Bertl hmm, anything in dmesg? 1269790339 M * oli nothing 1269790368 M * oli expect vxW with did lookup hidden devpts 1269790409 M * Bertl okay, so, depending on your kernel config and version, the guests might be waiting on an userspace process on the host 1269790414 M * Bertl (the vshelper) 1269790427 M * Bertl check on the host if there are any such processes 1269790496 M * oli kernel 2.6.31.1 with patch vs2.3.0.36.30.3 1269790505 M * oli vshelper non running on host 1269790542 M * Bertl util-vserver version? 1269790554 M * oli 0.30.216-pre2864 1269790598 M * Bertl okay, the reboot process in D state is waiting (uninterruptibly) for something 1269790609 M * Bertl the question now is, what is it waiting for? 1269790629 M * Bertl -f should relay to the kernel, which in turn will spawn the vshelper 1269790636 M * oli the only thing i did was executing vserver stop 1269790651 M * Bertl which should be fine 1269790658 M * oli ;) 1269790692 M * Bertl 31.1 is not the best choice though ... 1269790720 M * oli okay, is there a reason why? 1269790731 M * Bertl can you try if you can recreate the issue by creating a new guest, entering it and calling the reboot -d -f -i ? 1269790849 J * jrklein ~jrklein@2001:0:53aa:64c:0:539b:b97d:df43 1269790872 M * oli just a view secs 1269790927 M * Bertl np, take your time 1269791001 J * petzsch ~markus@p4FF4730C.dip.t-dialin.net 1269791266 M * oli hmm starting the guest hangs on /usr/lib/util-vserver/secure-mount --rbind -n -o dev ... 1269791328 M * oli what's bad with the 2.6.31.1 kernel? Better use 2.6.32.10? 1269791353 M * Bertl yes, we did a bunch of bug fixes after 2.6.31 1269791398 M * oli okay, first of all i will make an reset of the hostsystem and after that change to 2.6.32.10 1269791428 M * Bertl but I'm wondering, if you do not actually mean 2.6.33.1 ? 1269791446 M * oli sry I mean 2.6.33.1 typo 1269791459 M * Bertl okay, then stick with that, it should be fine 1269791484 M * Bertl i.e. we assume it is fine, if it doesn't work, it's a bug and we need to fix it :) 1269791518 M * oli hmm, how to track it down? 1269791531 M * Bertl so, some rbind hangs ... sounds like a device for a filesystem is busy? 1269791538 M * Bertl maybe some nfs involved? 1269791555 M * oli wi/o shows about 10% 1269791564 M * Bertl what does 'df' report? 1269791607 M * oli nothing special, what is the line you want to know? 1269791640 M * oli hmm jbd2/dm-0-8 need much cpu time 1269791723 M * Bertl try to issue a 'sync' 1269791791 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d 1269791807 M * Bertl wb bonbons! 1269791830 M * bonbons hey Bertl 1269791868 M * oli sync is running 1269791969 M * Bertl the question is, will it finish? 1269791980 M * Bertl also take a look at dmesg every now and then 1269792049 M * oli so this problem may come from ext4 ? 1269792080 M * Bertl it might be I/O or filesystem related 1269792089 M * Bertl but it might as well be the other way round 1269792278 M * oli okya 1269792296 M * oli so the problem is iO 1269792313 M * Bertl your current problem seems to be IO related 1269792321 M * oli okay i' 1269792331 M * oli ll search for i/o hevy tasks 1269792342 M * Bertl i.e. _something_ is waiting in the kernel (D) for something to happen (usually I/O) 1269792379 M * Bertl the question is, does the reboot trigger this issue? 1269792389 M * Bertl if so, we need to look there for a kernel bug 1269792398 M * Bertl (reboot is not something I/O intensive) 1269792567 Q * bonbons Quit: Leaving 1269792947 M * sleblanc After adapting my guest like indicated in the file Upstart, now the ssh server don't want start automatly. 1269792959 M * oli hmm it seems to be a kernel bug, because i tried to stop other guests and alle have the same issue 1269792972 M * oli load is always very high ~80 1269793047 M * sleblanc But, with /etc/init.d/ssh start, all seem good working. 1269793063 M * oli hmm sync just finished 1269793067 M * oli load is normalizing 1269793240 M * oli sry maybe no kernel problem then a userspace tool problem 1269793297 M * Bertl okay, what about the hanging reboot? 1269793426 M * oli gone after the heavy IO casuing process was killed 1269793435 M * Bertl excellent! 1269793450 M * Bertl yeah, I/O system in 2.6.32 and 33 IMHO still needs some work 1269793461 M * Bertl i.e. it easily gets pushed to its limits 1269793473 M * oli is there a possibility to find such processes easier? 1269793480 M * Bertl iotop 1269793565 M * oli does iotop show also the processes within guests? 1269793579 M * biz oli: create a script, eg. "viotop": 1269793583 M * biz setattr --watch /proc/vmstat 1269793583 M * biz exec chcontext --silent --ctx 1 iotop "$@" 1269793635 M * oli biz, thanks a lot 1269793841 M * biz np, and you can use this and look for processes in "D" state, like Bertl said: 1269793845 M * biz vps -eo pid,state,args --sort state 1269794248 M * oli maybe viotop helps me tracking down some other problems with exploding load 1269794354 M * Bertl you could even wrap up a patch to add it to util-vserver :) 1269794531 M * petzsch what should be in apps/init/style for a debian lenny vserver? i've switched between gentoo and debian in openVCP and it left me with style "gentoo" which of course leads to a non starting vserver 1269794560 M * petzsch before the "style" file didn't exist... just would wan't to know what to put in the image config file 1269794604 M * oli Bertl i discovered another strange thing: with patch patch-2.6.33.1-vs2.3.0.36.30.3.diff and delta-memcg-fix01.diff cgroup memory support stops working. In memory.limit_in_bytes all values are a strange value eg 982368746357632. In the guest there is no memory value applied. Even if you change the value nothing happens 1269794648 M * Bertl that sounds weird 1269794676 M * Bertl double check that you have the memory resource controller enabled 1269794807 M * oli check whether memory is in mountoption line for cgroup? 1269794910 M * Bertl primarily I'd check if it is enabled in the kernel 1269794924 M * Bertl but yeah, the mount part comes next 1269794983 M * oli with 2.6.33.1 and patch-2.6.33-vs2.3.0.36.30.3.diff with same kernel config this issue does not happen 1269795036 M * oli i'm also wondering why patch-2.6.33-vs2.3.0.36.30.3.diff and patch-2.6.33.1-vs2.3.0.36.30.3.diff do have very different size 1269795082 M * Bertl good point 1269795207 Q * balbir Ping timeout: 480 seconds 1269795237 M * Bertl it seems that the .33 patch contains a bunch of files which shouldn't really be there 1269795270 M * Bertl but I'll do a more in detail analysis on that to see if there are any functional differences 1269795310 M * oli ah okay, i'll check the patch-2.6.33-vs2.3.0.36.30.3.diff again 1269795345 M * Bertl you can test the .36.30.4 shortly 1269795354 M * oli okay i'll wait ;) 1269795374 M * Bertl should be up now 1269795721 J * balbir ~balbir@122.172.32.63 1269795788 M * Bertl escept for the removed hard limit scheduler in 2.6.33.1 I do not see any differences which might affect you 1269796103 M * Bertl *except 1269796113 M * Bertl clumsy fingers today ... 1269796327 M * oli why is the hard limit sheduler removed? 1269796340 M * oli is the lv scheduler back? 1269796434 M * Bertl no, we had some issues with the upstream patches, which turned out to be caused by a config option the developer assumed to be off 1269796458 M * Bertl there is now an alternative implementation, but I haven't had time to integrate that into the patchset yet 1269796481 M * oli alternative implementation mean? 1269796527 M * Bertl somebody from google (Paul Turner) did one 1269796567 M * Bertl but knowning that the issue we saw was caused by the config option (at least the developer says so :) 1269796590 M * Bertl you could patch back in the hardcfs 1269796613 M * Bertl (and make sure to disable user cfs) 1269796722 M * oli which config option is the one making troubles? 1269796757 M * oli delta-hardcfs-v4.diff is the patch for cgroup hard sheduling? 1269796826 M * oli hmm does not apply anymore on 2.6.33.1, ill wait for your new implementation ;) 1269796846 M * Bertl I uploaded the v5 version 1269796876 M * Bertl CONFIG_USER_SCHED=y is the problematic option 1269796976 M * oli okay much thanks 1269797011 M * Bertl but it won't apply cleanly, i.e. you need to pick the relevant parts 1269797179 M * oli i'll have a try 1269797226 M * oli thanks for your answers, i'll report if i get another time to the strange cgroup memory problematic 1269797234 M * oli off now, time to eat something 1269797345 M * Bertl cya 1269797396 Q * oli Quit: Leaving 1269799381 J * derjohn_mob ~aj@c152236.adsl.hansenet.de 1269800589 J * petzsch1 ~markus@p4FF476F7.dip.t-dialin.net 1269800742 Q * balbir Ping timeout: 480 seconds 1269800767 Q * sleblanc Quit: Leaving. 1269800942 Q * petzsch Ping timeout: 480 seconds 1269801251 J * balbir ~balbir@122.172.13.138 1269801436 Q * quasisane Read error: Operation timed out 1269801439 N * petzsch1 petzsch 1269801713 N * DoberMann[ZZZzzz] DoberMann[Flim] 1269801984 J * ntrs_ ~ntrs@77.29.0.99 1269802003 J * derjohn_foo ~aj@c195148.adsl.hansenet.de 1269802038 Q * ntrs Read error: Connection reset by peer 1269802420 Q * derjohn_mob Ping timeout: 480 seconds 1269802512 Q * balbir Ping timeout: 480 seconds 1269803041 J * balbir ~balbir@122.172.3.33 1269803319 M * petzsch has someone succesfully set up a gentoo guest? getting this error: http://paste.linux-vserver.org/14860 1269803354 M * Bertl Hollow: ping? 1269804473 M * petzsch switched from syslog-ng to metalog ... no error message this time 1269804533 M * Bertl ah, good! 1269805537 M * PowerKe petzsch: You could just comment the line 'file("/proc/kmsg");' in /etc/syslog-ng/syslog-ng.conf 1269805606 M * Bertl or alternatively enable the syslog virtualization 1269805689 M * Bertl off for dinner ... bbl 1269805693 N * Bertl Bertl_oO 1269806738 J * ghislain1 ~AQUEOS@adsl2.aqueos.com 1269806978 Q * ghislain Ping timeout: 480 seconds 1269807185 J * adamlis ~adamlis@d40-10.icpnet.pl 1269807260 M * adamlis should I ask questions here in english or could be polish too? 1269807303 M * petzsch english is prefered if you wan't more of us to be able to help you :-) 1269807318 M * adamlis thanks - then I'll try 1269807391 M * adamlis on host system there are few mountpoint below guest rootdir - I'd like to make guest aware of these mountpoints - so that running 'df' on guest would show them 1269807419 M * adamlis I've created /etc/vserver//apps/init/mtab file 1269807459 M * adamlis with content: /dev/hdv2 /path/to/where/guest/should/see/mountedpartition utf defaults 0 0 1269807484 M * adamlis this made "df" comman on guest correctly showing free space on this mountpoint 1269807497 M * adamlis and I'd like to ask if this is correct approach 1269807549 Q * hijacker_ Quit: Leaving 1269808115 N * Bertl_oO Bertl 1269808123 M * Bertl back now ... 1269808136 M * petzsch wb Bertl 1269808143 M * Bertl adamlis: yeah, sounds about right 1269808183 M * adamlis thanks 1269808199 M * adamlis reg. hdv1 hdv2 - can I freely choose these names? 1269808219 M * Bertl you can choose whatever you like there, nobody except userspaces looks at it 1269808264 M * adamlis ok - thanks 1269808651 Q * dna Quit: Verlassend 1269808703 N * DoberMann[Flim] DoberMann[ZZZzzz] 1269809337 M * bobnormal petzsch: i only run gentoo guests 1269809399 M * petzsch i personally prefer debian lenny but would like to offer my customers a little more choice 1269809409 M * bobnormal petzsch: haven't seen that error before, /proc/kmsg related errors have cropped up here before though 1269809422 M * bobnormal petzsch: i will check my openrc version, 1 sec 1269809443 M * petzsch i've followed this howto: http://www.openvcp.org/wiki/gentoo64bit-deutsch 1269809469 M * petzsch and used this stage archiv: http://bb.xnull.de/projects/gentoo/amd64/stage3-amd64-current.tar.bz2 1269809488 M * bobnormal petzsch: i use 0.4.3-r3 on mine 1269809542 M * petzsch i'm ok with running metalog and knowing about the workaround for syslog-ng... the rest is business of my users ;-) 1269809561 M * bobnormal petzch: unsure if it's related though, i think it's a host config issue, what initstyle are you using 1269809575 M * petzsch initstyle=gentoo 1269809704 M * bobnormal http://pastebin.com/VMKwM670 1269809717 M * bobnormal thats all the info i have off-hand 1269809745 M * bobnormal currently working to stupid deadline then offline for 3 days (now + 3 hours) so can't help much more sorry (also, wikipedia just getting admin-hack-attempts, obligated to report, argh ;) 1269809763 M * bobnormal good luck, email me if you need help walter at pratyeka 1269809770 Q * bobnormal Quit: . 1269809948 Q * ghislain1 Quit: Leaving. 1269813949 P * TeLLuS Bye 1269814503 Q * ntrs_ Ping timeout: 480 seconds 1269815038 M * petzsch off to get some sleep... good night 1269815146 Q * petzsch Quit: Leaving. 1269815728 J * adamlis_ ~adamlis@d40-10.icpnet.pl 1269816012 Q * adamlis Ping timeout: 480 seconds 1269816805 Q * jpic_ Ping timeout: 480 seconds 1269817507 Q * C14r Ping timeout: 480 seconds 1269819263 Q * adamlis_ Remote host closed the connection 1269819912 J * jpic ~jpic@perso.chocolatpistache.com 1269819932 P * jpic 1269819944 J * jpic ~jpic@perso.chocolatpistache.com