1288224211 Q * dowdle Remote host closed the connection 1288238950 Q * FireEgl Read error: Connection reset by peer 1288239935 J * FireEgl ~FireEgl@173-25-19-139.client.mchsi.com 1288240375 Q * balbir Ping timeout: 480 seconds 1288243157 J * balbir ~balbir@122.248.161.59 1288244743 J * ncopa ~ncopa@3.203.202.84.customer.cdi.no 1288245673 Q * julius Quit: Hackers of the world, unite! 1288247858 Q * aj__ Ping timeout: 480 seconds 1288248956 Q * FireEgl Ping timeout: 480 seconds 1288249307 J * aj__ ~aj@213.238.45.2 1288249696 J * thierryp ~thierry@lns-bzn-47f-62-147-212-202.adsl.proxad.net 1288251456 J * petzsch ~markus@dslb-092-078-151-023.pools.arcor-ip.net 1288251480 Q * balbir Ping timeout: 480 seconds 1288251921 Q * petzsch Quit: Leaving. 1288251969 N * Bertl_zZ Bertl 1288251973 M * Bertl morning folks! 1288252070 J * ein_bour ~chatzilla@mail2.tikalnetworks.com 1288252092 M * ein_bour hello. i installed a vserver again from scratch on a gentoo. 1288252097 M * ein_bour i build up a guest 1288252102 M * ein_bour and iam trying to start it 1288252110 M * ein_bour but it does not start 1288252120 M * ein_bour syslog and cron does not appear in the list 1288252125 M * ein_bour any hint? 1288252192 M * ein_bour here is the output of vserver guest start http://paste.linux-vserver.org/18268 1288252219 M * Bertl well, first, it probably starts fine, but, because you have no deamon running/active it stops when the last process dies 1288252231 M * ein_bour hello Bertl 1288252235 M * ein_bour we spoke yesterday 1288252251 M * ein_bour i installed vserver from scratch this time 1288252272 M * ein_bour i have a gentoo image that i have recovered and installed again vserver 1288252280 M * Bertl you can check/verify that by putting a 'sleep 60 &' in the local scripts (rc.local) 1288252295 M * Bertl which will keep the guest for 60 seconds 'running' 1288252310 M * ein_bour i have to do chroot? 1288252335 M * Bertl the idea with Linux-VServer without init process (note that you can also switch to the plain init style which will give you an init) 1288252353 M * Bertl is that only the processes requires (i.e. daemons) are running inside a guest 1288252375 M * Bertl if you enable cron or syslog or even sshd, the guest will keep running anyway 1288252387 M * Bertl I doubt that you really want a guest with no processes inside 1288252394 M * ein_bour of coruse 1288252399 J * barismetin ~barismeti@zanzibar.inria.fr 1288252399 M * Bertl (which you could also configure if you really want it) 1288252403 M * ein_bour i want a guest with processes that r running 1288252426 M * ein_bour i want to use vserver later on with openVCP to create vservers 1288252436 M * ein_bour i followed all the instructions 1288252448 M * ein_bour but the guest is starting without processes it seems 1288252466 M * ein_bour how do i enable processes to the guest? 1288252486 M * Bertl so, enable/add e.g. cron, syslog or sshd in the gentoo startup configuration (i.e. as service inside the guest) 1288252487 J * Romster ~romster@202.168.100.149.dynamic.rev.eftel.com 1288252531 M * Bertl note: I'm not a gentoo person, so you might need to ask somebody who knows how to do that (it is definitely gentoo userspace specific) 1288252576 M * Bertl if you want a quick guest for testing, create a debian guest, because it will be simple to install (done via network) and will contain/enable basic services 1288252608 M * Bertl for details on gentoo guests, best have a chat with Hollow, he should know all the details 1288252630 M * ein_bour is he around now? 1288252810 M * Bertl he might not be 'around' but he will definitely read and answer your questions sooner or later 1288252840 Q * ntrs Ping timeout: 480 seconds 1288252908 M * ein_bour Bertl: how can i chroot to my guest 1288252910 M * ein_bour ? 1288252914 M * ein_bour what is the command? 1288253175 M * Bertl chroot ? 1288253188 M * ein_bour yes 1288253199 M * Bertl chroot is the command to chroot to your guest 1288253203 M * ein_bour i want to chroot to the guest 1288253210 M * Bertl then do it :) 1288253211 M * ein_bour i meant the syntax 1288253227 M * Bertl 'man chroot' 1288253243 M * Bertl but basically 'chroot /path/to/guest' 1288253262 J * ntrs ~ntrs@vault08.rosehosting.com 1288253565 M * ein_bour thanks 1288253677 J * balbir ~balbir@122.248.161.59 1288253755 M * ein_bour i tried to emerge syslog-ng to my guest path 1288253769 M * ein_bour i chroot to my guest, rc-update add syslog-ng 1288253771 J * manana ~mayday090@84.17.25.149 1288253775 M * ein_bour but i cannot start it 1288253787 M * ein_bour missing libssl.so 1288253838 M * ein_bour can i copy it to the guest? 1288254068 M * swenTjuln ein_bour: you've broken chroot environment 1288254088 M * ein_bour how do i fix it? 1288254099 M * swenTjuln Bertl: i've a problem regarding ulimit 1288254175 M * swenTjuln ein_bour: have you followed any guide how to install gentoo? 1288254198 M * swenTjuln ein_bour: with gentoo you must really know what you're doing 1288254203 M * ein_bour i have an image for ubuntu 1288254206 M * ein_bour sorry 1288254207 M * ein_bour gentoo 1288254220 M * swenTjuln well....create image for ubuntu ;) 1288254227 M * swenTjuln its really easy 1288254230 M * ein_bour lol :) 1288254236 M * ein_bour i need tow work with gentoo 1288254245 M * ein_bour i installed the vserver, guest 1288254249 M * ein_bour but i cannot start it :S 1288254255 M * ein_bour missing processes 1288254274 M * ein_bour i emerged syslog to the guest 1288254282 M * ein_bour can not start it now due to missing libssl 1288254293 M * ein_bour i updated rc 1288254295 M * swenTjuln ein_bour: then you should really follow http://www.gentoo.org/doc/en/handbook/handbook-x86.xml . It's detailed step-by-stepo installation. You can leave out all low level stuff 1288254297 M * ein_bour i need to start it now 1288254323 M * swenTjuln emerge should pull in all dependencies 1288254343 M * swenTjuln so you've done something very wrong 1288254421 M * swenTjuln ein_bour: what did you do so far (installing gentoo guest)? 1288254428 M * ein_bour yes 1288254434 M * ein_bour tried to start it 1288254439 M * ein_bour it starts but stops 1288254440 M * swenTjuln what? 1288254449 M * swenTjuln how did you create guest? 1288254450 M * ein_bour no syslog nor cron running 1288254469 M * ein_bour following the link http://www.gentoo.org/proj/en/vps/vserver-howto.xml 1288254473 M * swenTjuln ein_bour: did you install _guest_? 1288254482 M * ein_bour yes 1288254485 M * ein_bour following the link 1288254486 M * ein_bour stage 3 1288254507 M * ein_bour i can see the folder under /vservers 1288254507 M * urbee ein_bour: use a template 1288254517 M * ein_bour which one? 1288254537 M * urbee w8 1288254551 M * swenTjuln ein_bour: what is an output of vserver-info ? 1288254595 M * ein_bour vserver-info: http://paste.linux-vserver.org/18269 1288254621 M * swenTjuln Bertl: is vserver-util somehow forceinf ulimit? 1288254719 M * swenTjuln ein_bour: so you've downloaded template? 1288254731 M * ein_bour template? 1288254732 M * ein_bour no 1288254805 J * FireEgl ~FireEgl@173-25-19-139.client.mchsi.com 1288254825 M * urbee ein_bour: ftp://ftp.pld-linux.org/people/hawk/vserver-templates/ 1288254829 M * urbee here u have some templates 1288254834 M * urbee maybe abit outdated 1288254851 M * ein_bour gentoo? 1288254854 M * urbee i dont know where my gentoo is, but it was a few years old anyway 1288254854 M * urbee :p 1288254877 M * swenTjuln ein_bour: so you've downloaded stage3 image? 1288254886 M * ein_bour yes i have stage3 image 1288254899 M * ein_bour with the stage3 i have built the guest 1288254971 M * swenTjuln ok....then you've created vserver guest using: vserver myguest build command , as described in howto 1288254979 M * ein_bour yes 1288254992 M * ein_bour as described in the link 1288255077 M * swenTjuln output of: ls /etc/vservers/* -l 1288255217 M * ein_bour http://paste.linux-vserver.org/18270 1288255238 M * ein_bour cache and run seem to be broken though 1288255323 M * swenTjuln ein_bour: that is OK, becouse container is not running 1288255371 M * ein_bour just to let u know 1288255386 M * ein_bour swenTjuln: 1288255397 M * ein_bour i emerged syslog-nd into the guest 1288255400 M * ein_bour added to rc 1288255404 M * ein_bour starting now 1288255410 M * ein_bour complained about missing libs 1288255419 M * ein_bour i copied them to the guest 1288255429 M * ein_bour now complainingabout /proc/kmsg 1288255438 M * swenTjuln OH! 1288255440 M * swenTjuln :D 1288255449 M * ein_bour Error opening file for reading; filename='/proc/kmsg', error='Operation not permitted (1)' 1288255474 M * Bertl swenTjuln: what do you mean by 'forceinf ulimit'? 1288255496 M * swenTjuln Bertl: forceing ulimit :D sry for misspell 1288255505 M * Bertl ein_bour: just comment out the /proc/kmsg in the syslog config, or enable the syslog virtualization for the guest 1288255540 M * ein_bour here is the output of the vserver guest start http://paste.linux-vserver.org/18271 1288255549 M * ein_bour i will try that now 1288255553 M * swenTjuln on vserver host i have ulimit -n 100000, then I start a guest, and within guest i've ulimit -n 1024 1288255586 M * Bertl yes, that is expected, guests have independant limits 1288255603 M * Bertl (you can configure them in the guest config, see 'Flower Page') 1288255619 J * petzsch ~markus@dslb-092-078-151-023.pools.arcor-ip.net 1288255630 M * swenTjuln ein_bour: you should remove "pipe("/proc/kmsg"); " part from "source src { unix-stream("/dev/log"); internal(); pipe("/proc/kmsg"); }" in etc/syslog-ng/syslog-ng.conf in guest 1288255648 M * swenTjuln Bertl: so 1024 is ddefault? 1288255666 M * ein_bour i commented all the source src 1288255671 M * ein_bour and not it starts 1288255673 M * ein_bour now* 1288255694 M * ein_bour is it ok to comment all the part? 1288255731 M * swenTjuln no 1288255749 M * Bertl swenTjuln: it uses the kernel defaults for a guest, so yes, if not set, this is likely the default 1288255763 M * swenTjuln ein_bour: instead of "source src { unix-stream("/dev/log"); internal(); pipe("/proc/kmsg"); }" put "source src { unix-stream("/dev/log"); internal(); }" 1288255778 M * swenTjuln Bertl: thank you! 1288255789 M * ein_bour instead of pipe its file 1288255798 M * ein_bour i just commented the line of /proc/kmsg 1288255802 M * ein_bour and it starts now 1288255825 M * swenTjuln ein_bour: 1288255828 M * swenTjuln ein_bour: good 1288255837 M * ein_bour i have an entry in vserver-stat: 1253 2 9.4M 2.1M 0m00s10 0m00s14 0m53s80 guest 1288255843 M * ein_bour so i guess its up 1288255847 M * swenTjuln now you should be able to enter with: vserver guest enter 1288255854 M * ein_bour what about cron and other processes? 1288255868 M * swenTjuln ein_bour: you must configure them too 1288255874 M * ein_bour i am in yeah 1288255878 M * swenTjuln one-by-one 1288255880 M * swenTjuln :D 1288255884 M * ein_bour :D 1288255886 M * ein_bour lovely 1288255892 M * swenTjuln ein_bour: happy hacking ;) 1288255936 M * ein_bour now its running, i am running vesync --sync guest 1288255948 M * ein_bour update the portage tree 1288255957 M * ein_bour enabling emerge i hope in the guest 1288256010 M * ein_bour i chroot in the guest 1288256019 M * ein_bour and run emerge --sync 1288256029 M * ein_bour its complaining that /proc is not mounted 1288256138 M * Bertl did you run the scripts on the host (to setup util-vserver)? 1288256148 M * Bertl (they are usually part of the startup scripts) 1288256205 M * ein_bour i just emerged util-vserver 1288256210 M * ein_bour what scripts? 1288256298 M * Bertl emerged where? 1288256327 M * ein_bour to the host 1288256372 M * Bertl well, if you didn't reboot after that, you definitely want to run those scripts (but maybe they got already executed and you have a different problem) 1288256390 J * kir ~kir@swsoft-msk-nat.sw.ru 1288256422 M * Bertl ah, you chrooted into the guest now? 1288256444 M * Bertl don't do that when it is running, use 'vserver enter' instead 1288256477 M * ein_bour ok 1288256482 J * harobed ~harobed@pda57-1-82-231-115-1.fbx.proxad.net 1288256510 M * ein_bour u r right 1288256518 M * ein_bour no /proc message 1288256542 M * ein_bour but not it seems network issue !!! getaddrinfo failed for 'rsync.gentoo.org': [Errno -2] Name or service not known 1288256575 M * ein_bour i checked ifconfig 1288256581 M * ein_bour i have an ip 1288256583 M * ein_bour but no ping 1288256794 M * ein_bour fixxed 1288256866 M * swenTjuln ein_bour: fix /etc/resolv.conf 1288256878 M * ein_bour yes i did 1288256897 M * ein_bour i am reading the link u posted swenTjuln 1288256901 M * ein_bour fixed the /proc 1288256907 M * ein_bour and resolv.conf 1288256919 M * ein_bour i am running emerge --sync under the guest now 1288256931 M * ein_bour its so complicated though 1288257032 M * swenTjuln ein_bour: it really isn't. It's just that you have to know a little about linux,gentoo specific and vserver specific issues 1288257072 M * renihs whats complicated in a emerge --sync? :) 1288257109 M * swenTjuln renihs: nothin' much until it works as expected ;) 1288257110 M * ein_bour renihs: what r u talking about 1288257115 M * ein_bour its not compliacated 1288257129 M * ein_bour the whole process of installing vserver, setting up guest 1288257134 M * ein_bour is complicated 1288257172 M * renihs it is? 1288257210 M * renihs i tend to untar a stage, adjust the xid, make some links and start it 1288257210 M * swenTjuln ein_bour: I think you're missing knowledge of what jail, chroot, containter is. It would clearify alot for you 1288257213 M * ein_bour for u its not i guess 1288257223 M * renihs its complicated the first time :) 1288257231 M * renihs since you dont see the whole thing at that stage 1288257248 M * ein_bour i agree with u on that 1288257250 M * renihs but afterwards, you will admit, its not actually complicated :) 1288257261 M * renihs after doing it a couple times sucessfully :) 1288257277 M * ein_bour thats it 1288257280 M * ein_bour couple of times 1288257395 Q * ntrs Ping timeout: 480 seconds 1288257669 Q * kir Remote host closed the connection 1288257882 J * ntrs ~ntrs@vault08.rosehosting.com 1288257949 J * kir ~kir@swsoft-msk-nat.sw.ru 1288258437 M * swenTjuln Bertl: I still have that ulimit problem 1288258454 M * swenTjuln seems like i've just dropped limits shown by vlimit 1288258486 M * Bertl vlimit = rlimit, ulimits are separate 1288258493 M * fLoo lol Bertl 1288258503 M * fLoo you finally got the util-vserver onto the mainpage 1288258506 M * fLoo congratz ;) 1288258513 M * swenTjuln oh... 1288258539 M * swenTjuln .....my bad 1288258592 M * Bertl np 1288259094 M * swenTjuln Bertl: are ulimits not working with util-vserver 0.30.215 = 1288259096 M * swenTjuln ? 1288259113 M * Bertl depends, what kernel? 1288259138 M * swenTjuln 2.6.32-25-vserver Ubuntu from PPA 1288259152 M * Bertl nope, too old (util-vserver) 1288259160 M * swenTjuln i see 1288259171 M * Bertl 0.30.215 won't even get the guest isolation right with that kernel 1288259191 J * ktwilight__ ~keliew@91.176.24.245 1288259367 M * ntrs daniel_hozac: rpm.hozac.com seems to be down. any info? 1288259483 Q * ktwilight_ Ping timeout: 480 seconds 1288259564 M * swenTjuln Bertl: can I update util-vserver without restarting guests? 1288259592 M * Bertl yes, but you need to restart them to get the isolation right 1288259599 M * swenTjuln Bertl: and which version would you recommend for my kernel? 1288259635 M * swenTjuln great. So i0ll restart them later :D 1288260080 Q * petzsch Quit: Leaving. 1288260096 M * swenTjuln Bertl: would I be right if i'd assume util-vserver-0.30.216-pre2908 is currently prefered version? 1288260154 M * swenTjuln ...although it's pre2914 is latest 1288260192 M * Bertl how so? usually the latest pre is the advised one 1288260240 M * swenTjuln Bertl: it's in a topic :D 1288260402 F * ChanServ +o Bertl 1288260407 T * Bertl http://linux-vserver.org/ |stable 2.2.0.7, exp 2.3.0.36.37, grsec 2.3.0.36.28|util-vserver-0.30.216-pre2914| He who asks a question is a fool for a minute; he who doesn't ask is a fool for a lifetime -- share the gained knowledge on the Wiki, and we forget about the minute. 1288260792 J * stYg ~styg@adsl-dyn-235.95-102-83.t-com.sk 1288260828 Q * harobed Ping timeout: 480 seconds 1288260884 Q * renihs Quit: narf 1288261300 P * kir Leaving. 1288262187 J * renihs ~lemming@83-65-34-34.arsenal.xdsl-line.inode.at 1288262968 J * thierryp_ ~thierry@lns-bzn-47f-62-147-212-202.adsl.proxad.net 1288262969 Q * thierryp Read error: Connection reset by peer 1288263181 Q * harry Ping timeout: 480 seconds 1288263267 J * petzsch ~markus@dfn212.rz.tu-ilmenau.de 1288263602 Q * petzsch Quit: Leaving. 1288263834 Q * Piet Quit: Piet 1288264574 M * ein_bour how do i delete a guest? 1288264630 Q * balbir Ping timeout: 480 seconds 1288264807 M * Bertl vserver --help 1288267040 Q * stYg Ping timeout: 480 seconds 1288267794 J * harobed ~harobed@pda57-1-82-231-115-1.fbx.proxad.net 1288267957 M * swenTjuln Bertl: still can't get ulimit -n > 1024 1288267998 M * swenTjuln Kernel: 2.6.32-24-vserver; Utils: 0.30.216-pre2914 1288268064 M * swenTjuln It does work if I do: vserver enter, but when I ssh to the guest it's still 1024 :( 1288268078 M * swenTjuln any thougths 1288268354 M * Bertl yes, most likely pam or similar sets the limit for you 1288268421 M * Bertl i.e. check /etc/pam* and /etc/security/limits.conf 1288268443 M * Bertl (also note that this is a soft limit, hard limits require -H) 1288268504 J * BenG ~bengreen@cpc6-aztw22-2-0-cust100.aztw.cable.virginmedia.com 1288269077 J * thierryp ~thierry@lns-bzn-47f-62-147-212-202.adsl.proxad.net 1288269168 Q * BenG Ping timeout: 480 seconds 1288269231 Q * thierryp_ Ping timeout: 480 seconds 1288269471 J * petzsch ~markus@dfn212.rz.tu-ilmenau.de 1288270270 M * ein_bour i installed openVCP. 1288270288 M * ein_bour openVCP works along with vserver 1288270307 M * ein_bour i am trying to create vservers from the web interface but i am getting this error 1288270363 M * ein_bour 'vserver ... suexec' is supported for running vservers only; aborting... 1288270363 M * ein_bour any hint? 1288270416 M * ein_bour i checl the dir/vservers 1288270416 M * ein_bour and the vserver i create is not created there 1288270416 M * ein_bour but i can found it under /etc/vservers 1288270586 M * petzsch ein_bour: please ask your openVCP related questions in the #openvcp channel on this irc server 1288270610 M * ein_bour ok, thanks 1288270628 M * ein_bour since openVCP works alongside vserver 1288270634 M * ein_bour i thought someone could help 1288270660 M * ein_bour the error i am getting is related to vserver 1288270660 M * ein_bour 'vserver ... suexec' is supported for running vservers only; aborting... 1288270756 Q * ntrs Ping timeout: 480 seconds 1288270817 Q * petzsch Quit: Leaving. 1288270947 M * swenTjuln strangest thing: if I set ulimit in /etc/security/limits.conf it doesn't work. But if I commtent out "session required pam_limits.so" then it works 1288271220 J * ntrs ~ntrs@vault08.rosehosting.com 1288271911 M * Bertl swenTjuln: just means that pam_limits uses a different file 1288271918 M * Bertl (or no file at all :) 1288272031 M * ein_bour Bertl: any hint regarding 'vserver ... suexec' is supported for running vservers only; aborting... 1288272060 M * Bertl well, it is supported for running guests only :) 1288272084 M * swenTjuln Bertl: I've failed to see "NOTE: group and wildcard limits are not applied to root." :\ 1288272100 M * Bertl ah, that explains 1288272113 M * swenTjuln yup 1288272118 M * ein_bour Bertl: what do u mean? 1288272179 M * Bertl I mean, if your guest is not running, then suexec won't work (on it) 1288272187 M * ein_bour i can create vservers in openVCP? 1288272187 M * ein_bour can't* 1288272237 M * ein_bour but the guest is running 1288272259 M * ein_bour 1253 3 11.5M 2.8M 0m00s15 0m00s15 20m39s42 guest 1288272280 M * ein_bour openVCP recognizes the guest I build with vserver 1288272390 M * ein_bour but trying to create new vservers 1288272390 M * ein_bour i fail 1288272390 M * ein_bour they fail to start 1288272390 M * ein_bour folders will be created in /etc/vservers 1288272390 M * ein_bour but not in /vserver 1288272437 M * daniel_hozac sounds like OpenVCP isn't creating them correctly. 1288272477 M * Bertl ein_bour: check with fLoo ... 1288272509 M * ein_bour i am contacted him on openvcp channel 1288272512 M * Bertl daniel_hozac: got a minute or two? 1288272513 M * ein_bour still did not reply 1288272551 M * daniel_hozac Bertl: sure. what's up? 1288272552 J * balbir ~balbir@122.172.32.146 1288272554 M * Bertl ein_bour: well, IRC is like chat but with some delay 1288272555 M * ein_bour daniel_hozac: it seems yeah 1288272555 M * ein_bour it creates only in one place 1288272555 M * ein_bour should create them in /vservers also 1288272559 M * ein_bour what i am missing? 1288272574 M * Bertl ein_bour: openvcp related, wrong channel here to ask 1288272617 M * Bertl daniel_hozac: just a few things and planned changes on my side ... so that we can synchronize 1288272642 M * ein_bour Bertl: i know yeah. 1288272658 M * Bertl daniel_hozac: I'm going to remove (at least the majority) of the scheduler code 1288272680 M * Bertl (which basically is inactive since 2.6.26+) 1288272707 M * daniel_hozac right 1288272710 M * Bertl the question is, how do we handle the priority bias (which is per cpu and integrated into the scheduler interface 1288272727 M * Bertl I see a bunch of options there: 1288272742 M * ein_bour i joined openvcp and i am asking there 1288272749 M * Bertl good idea 1288272760 M * Bertl - leave the various scheduler syscalls as dummies, only extracting the priority bias 1288272786 M * Bertl - have the kernel block/reject all scheduler syscalls (except for the priority bias) 1288272818 M * Bertl - remove the entire syscall commands and structures (which I would really love to do) and have a different priority bias syscall command 1288272874 M * Bertl now, what I don't know is how much work the third option would be on your side, and what guest startup implications this would bring for 'older' tools? 1288272910 M * daniel_hozac it shouldn't be too much work. 1288272933 M * daniel_hozac and honestly, i don't think the priority bias is used all that much. 1288272947 M * daniel_hozac we already have nice which is probably more familiar to people. 1288272962 M * Bertl I'm not so much worried about that, I'm more worried about taking the scheduler syscalls away 1288272983 M * Bertl i.e. what will happen to a guest with scheduler parameters configured? 1288273008 M * Bertl (not that they apply atm) 1288273032 M * daniel_hozac it would fail to start. 1288273047 M * daniel_hozac which is probably a good thing anyway, so people don't think they are limited when they're really not. 1288273068 M * Bertl with a somewhat meaningful message to pinpoint the issue? 1288273094 M * daniel_hozac it'd be a ENOSYS from vsched, so should be fairly obvious... 1288273116 M * Bertl okay, that sounds fine to me, so I take it that this would be your preferred option as well? 1288273164 M * daniel_hozac yeah, sounds like the best way to do. 1288273168 M * daniel_hozac s/do/go/ 1288273205 M * Bertl regarding the priority bias, I'd drop that immediately, but I couldn't figure out a way with cgroups to priorize cpus differently for a guest 1288273238 M * Bertl and it is somewhat helpful to have a guest mostly use one cpu over the other, but not exclusively 1288273393 M * Bertl okay, another thing is the fact that vserver-stat still doesn't work under certain circumstances (in the latest pre) 1288273460 M * Bertl vc_rlimit_stat(RLIMIT_AS): Invalid argument -> no output 1288273482 M * Bertl is that something we need to address in the kernel by disallowing something? 1288273539 M * daniel_hozac with what util-vserver version on what kernel? 1288273573 M * Bertl 0.30.216-pre2914 on 2.6.36-vs2.3.0.36.37 1288273584 M * daniel_hozac without memory cgroups, i presume? 1288273601 M * Bertl # CONFIG_VSERVER_LEGACY_MEM is not set 1288273610 M * Bertl and no cgroup configured 1288273620 M * daniel_hozac right. 1288273637 M * daniel_hozac that's expected, but i suppose we can still show the other data... 1288273646 M * daniel_hozac or use the old hack of iterating the processes. 1288273670 M * Bertl that would be really helpful I guess, because I think this is the 'default' setup for most folks who start using Linux-VServer 1288273689 M * Bertl (at least in the future, when the LEGACY_MEM is gone) 1288273709 M * daniel_hozac do you know at which API version we dropped support for RLIMIT_AS? 1288273720 M * Bertl give me a second to check 1288273832 M * Bertl 0x00020305 , kernel 2.6.33 was the last one with AS limit 1288273862 M * daniel_hozac was there a bump there? 1288273866 M * Bertl kernel 2.6.34 already has 0x00020306 1288273872 M * daniel_hozac good. 1288273873 M * daniel_hozac okay. 1288273906 M * Bertl shall I bump the VCI again when removing the scheduler syscalls? 1288273926 M * daniel_hozac ideally. 1288273936 M * Bertl okay, np 1288274030 M * Bertl I presume we should keep the scheduler command structures around at least for the headers, or do you handle that yourself in util-vserver (in future kernel header updates)? 1288274070 M * daniel_hozac yeah, i keep old structures around myself. 1288274078 M * Bertl okay, great! 1288274083 M * daniel_hozac we can remove them from the kernel. 1288274104 M * Bertl finally, any things you need to be added/changed in the kernel from the util-vserver PoV? 1288274144 M * daniel_hozac not that i can think of, but i'll admit i've been pretty absent lately. 1288274278 M * Bertl ah, well, took the mainline folks about half a year to come up with a new patch, so everything is relative :) 1288274312 M * daniel_hozac hehe 1288274525 Q * manana Remote host closed the connection 1288276577 J * BenG ~bengreen@cpc6-aztw22-2-0-cust100.aztw.cable.virginmedia.com 1288277384 J * dowdle ~dowdle@scott.coe.montana.edu 1288277717 Q * ein_bour Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100915193946] 1288278050 J * alpha_one_x86 ~kvirc@161.30.17.95.dynamic.jazztel.es 1288278079 M * alpha_one_x86 With the lastest kernel gentoo source, I have lot of error with vserver + grsec: 1288278097 Q * ncopa Quit: Ex-Chat 1288278123 M * alpha_one_x86 http://pastebin.com/trmwwFr7 1288278363 M * Bertl well, probably gentoo has some patches on that kernel as well 1288278384 M * Bertl I'd suggest to go for a vanilla kernel with Linux-VServer (and if you want grsec) 1288278388 M * alpha_one_x86 Yes, to fix some security bug 1288278405 M * Bertl security bugs not addressed in the latest vanilla kernel? 1288278409 M * alpha_one_x86 vanilla + same security bug will do the same 1288278436 M * alpha_one_x86 I have only this error since gentoo have include security fix 1288278447 M * Bertl what kernel are we talking about? 1288278464 M * alpha_one_x86 gentoo kernel 1288278481 M * Bertl yeah, I got that from the context :) but which version? 1288278494 M * alpha_one_x86 2.6.32-r20 1288278512 M * Bertl so, that is quite old, go for 2.6.36 1288278552 J * ryker ~ryker@199.117.46.145 1288278581 M * Bertl (or if you really want to use 2.6.32, find somebody to adapt the patches (or adapt them yourself) 1288278587 M * ryker can anyone tell me where I can find a beecrypt-devel rpm? Google is failing me. :/ 1288278615 M * Bertl http://fr2.rpmfind.net/linux/rpm2html/search.php?query=lib64beecrypt-devel%28x86-64%29 1288278668 M * daniel_hozac what operating system? 1288278670 M * ryker thx Bertl. I only see 3 for mandriva. I could have sworn I found one before for CentOS, but I can't find it now 1288278681 M * daniel_hozac just yum install beecrypt-devel? 1288278684 M * daniel_hozac it's in base. 1288278699 Q * BenG Quit: I Leave 1288278700 M * ryker hi daniel_hozac. you're repo is down. at least it is for me 1288278718 M * ryker and I couldn't find it in a mirror 1288278754 M * daniel_hozac yes, my firewall's PSU is broken. should be taken care of in the next few days. 1288278793 M * ryker anywhere else you know of I can get that rpm from for CentOS? 1288278809 M * daniel_hozac which one? 1288278818 M * ryker beecrypt-devel 1288278830 M * daniel_hozac it's part of CentOS. 1288278838 M * alpha_one_x86 .rej is like normal patch? 1288278849 M * daniel_hozac any CentOS mirror has it. 1288278854 M * Bertl ryker: (means: any CentOS mirror will do) 1288278867 M * ryker hmm. i didn't think it was. i thought I tried it. let me try again. 1288278878 M * Bertl alpha_one_x86: depends, but usually yes 1288278892 M * alpha_one_x86 To fix the patch I need apply the .rej manually? ->> then I do that's 1288278900 M * ryker lol. ok. thx. it worked. sry to waste your time. 1288278923 M * Bertl np, have fun! 1288278934 M * alpha_one_x86 I understand why it have not work.... 1288279016 M * alpha_one_x86 The patch: 1288279016 M * alpha_one_x86 http://pastebin.com/XcN5uqnB 1288279058 M * alpha_one_x86 My kernel source: 1288279059 M * alpha_one_x86 http://pastebin.com/88Gi6pfE 1288279064 M * alpha_one_x86 How solve it? 1288279159 M * ryker daniel_hozac: any chance the vserver utils will work with the latest vserver kernels any time soon? ie. vserver-stat and vlimit, etc. 1288279171 M * daniel_hozac they already do.... 1288279195 M * ryker with the 2.6.35 kernel patches? 1288279203 M * ryker it doesn't work for me 1288279234 M * ryker i get vc_rlimit_stat(RLIMIT_AS): Invalid argument when I run vserver-stat 1288279252 M * ryker and it shows an empty list of running instances 1288279255 M * daniel_hozac do you have memory cgroups enabled in your kernel? 1288279269 M * ryker yes, should i disable cgroups? 1288279330 M * daniel_hozac no. 1288279343 M * daniel_hozac you need them and a recent util-vserver. 1288279362 M * ryker we were happy with the stable vserver kernel and everything worked great, but we had to compile our own kernel to use drives only available in 2.6.35, but now resource limiting doesn't work right for us and stuff like vserver-stat 1288279389 M * ryker daniel_hozac: I just used rpmbuild to build util-vserver pre2914 1288279399 M * ryker is there newer? 1288279432 M * daniel_hozac 2918, but that's new enough. 1288279459 J * petzsch ~markus@dslb-092-078-151-023.pools.arcor-ip.net 1288279471 M * daniel_hozac what does grep memory /proc/cgroups return? 1288279554 M * ryker need a couple minutes to check. i just rebooted the server after installing the new utils. Probably wasn't necessary, but I figured it couldn't hurt. 1288279891 M * ryker daniel_hozac: the server must not have had grub configured properly, because it hasn't come back up. It will probably be a while before I can check /proc/cgroups, but I see the cgroups wiki page now: http://linux-vserver.org/util-vserver:Cgroups 1288279906 M * ryker i'll take a look at that when I get the server running again. 1288279916 M * ryker thx for your help. 1288280288 M * Bertl no remote/serial console? 1288280321 M * ryker Bertl: yes, i'm working away from the office right now though, so i have a co-worker fixing it for me. 1288280338 Q * alpha_one_x86 Quit: KVIrc Equilibrium 4.1.1, revision: 5000, sources date: 20100905, built on: 2010-10-25 14:39:57 UTC http://www.kvirc.net/ 1288280417 M * ryker daniel_hozac: ok. finally got it. 1288280423 M * ryker cat /proc/cgroups 1288280423 M * ryker #subsys_name hierarchy num_cgroups enabled 1288280423 M * ryker cpuset 1 1 1 1288280423 M * ryker ns 0 1 1 1288280423 M * ryker cpu 1 1 1 1288280426 M * ryker cpuacct 1 1 1 1288280428 M * ryker devices 1 1 1 1288280431 M * ryker freezer 1 1 1 1288280588 M * ryker daniel_hozac: so, is there something I need to change to get vserver-stat to list the running instances with this newer kernel? I'm using 2.6.35-vs2.3.0.36.31 1288280633 M * daniel_hozac yes, enable the memory cgroup. 1288280945 M * ryker daniel_hozac: is this what I need to do? 'mount -t cgroup -ocpu,cpuset,memory none /dev/cgroup' 1288280962 M * ryker I get an error that special device none doesn't exist. 1288281108 M * daniel_hozac no. 1288281119 M * daniel_hozac in your kernel, enable the memory cgroup. 1288281131 M * daniel_hozac CONFIG_CGROUP_MEM_RES_CTRL IIRC. 1288281157 M * ryker ah, ok. 1288281166 M * ryker thx. off to build a new kernel 1288281766 J * bonbons ~bonbons@2001:a18:1:1402:2c0:9fff:fe2d:39d 1288282430 Q * harobed Ping timeout: 480 seconds 1288282662 Q * barismetin Remote host closed the connection 1288282846 M * ryker is there any reason not to enable the 'Legacy Memory Limits' option in the kernel? I'm wondering if this will let me use openvcp to manage resource limits, since it isn't cgroup aware yet. 1288282897 M * Bertl forget it, it's just a fake to allow older tools to work 1288282930 M * Bertl (check the help, the old accounting is gone for good) 1288282984 M * ryker ok. thx Bertl 1288283433 J * hijacker_ ~hijacker@87-126-142-51.btc-net.bg 1288284732 Q * aj__ Ping timeout: 480 seconds 1288285504 J * dna ~dna@dslb-094-222-214-138.pools.arcor-ip.net 1288285563 J * BenG ~bengreen@cpc6-aztw22-2-0-cust100.aztw.cable.virginmedia.com 1288285840 Q * BenG 1288288400 T * Bertl http://linux-vserver.org/ |stable 2.2.0.7, exp 2.3.0.36.38, grsec 2.3.0.36.28|util-vserver-0.30.216-pre2914| He who asks a question is a fool for a minute; he who doesn't ask is a fool for a lifetime -- share the gained knowledge on the Wiki, and we forget about the minute. 1288288962 J * Piet ~Piet__@04ZAAAU2N.tor-irc.dnsbl.oftc.net 1288289373 M * Bertl off for now .. bbl 1288289379 N * Bertl Bertl_oO 1288290052 J * aj__ aj@tmo-061-210.customers.d1-online.com 1288291786 J * harobed ~harobed@arl57-1-82-231-110-14.fbx.proxad.net 1288292224 J * imcsk8 ~ichavero@201.174.32.230 1288292397 J * harry ~harry@d51A461B4.access.telenet.be 1288292432 Q * harry 1288292452 J * harry ~harry@d51A461B4.access.telenet.be 1288293847 Q * thierryp Remote host closed the connection 1288294292 Q * petzsch Quit: Leaving. 1288294818 Q * aj__ Ping timeout: 480 seconds 1288295392 J * dna_ ~dna@dslb-092-078-097-164.pools.arcor-ip.net 1288295638 J * dna__ ~dna@dslb-094-222-217-037.pools.arcor-ip.net 1288295823 Q * dna Ping timeout: 480 seconds 1288296063 Q * dna_ Ping timeout: 480 seconds 1288296102 J * aj__ aj@tmo-040-157.customers.d1-online.com 1288296157 J * dna_ ~dna@dslb-094-222-217-037.pools.arcor-ip.net 1288296248 J * petzsch ~markus@dslb-092-078-151-023.pools.arcor-ip.net 1288296576 Q * dna__ Ping timeout: 480 seconds 1288296966 Q * ntrs Ping timeout: 480 seconds 1288297451 J * ntrs ~ntrs@vault08.rosehosting.com 1288297786 Q * aj__ Ping timeout: 480 seconds 1288298375 Q * imcsk8 Ping timeout: 480 seconds 1288298386 J * imcsk8 ~ichavero@201.174.32.230 1288298500 N * Bertl_oO Bertl 1288298504 M * Bertl back now ... 1288298861 Q * Chlorek Ping timeout: 480 seconds 1288299004 J * Chlorek ~cokolwiek@c.sed.pl 1288299199 Q * hijacker_ Quit: Leaving 1288299229 Q * bonbons Quit: Leaving 1288299411 Q * dna_ Quit: Verlassend 1288299505 J * cuba33ci_ ~cuba33ci@111-240-210-181.dynamic.hinet.net 1288299855 Q * cuba33ci Ping timeout: 480 seconds 1288300139 M * ryker on centos5.5, rsyslog hangs when stopping a guest. I fixed this problem with syslog-ng before by commenting out the section about stopping the kernel logger. rsyslog doesn't have anything about kernel logging in the init file. any suggestions? 1288300496 Q * imcsk8 Quit: Leaving 1288300625 Q * ntrs Ping timeout: 480 seconds 1288300923 M * Bertl 'hangs' means? 1288300947 J * manana ~mayday090@84.17.25.149 1288301089 J * ntrs ~ntrs@vault08.rosehosting.com 1288301258 Q * harobed Ping timeout: 480 seconds 1288301971 M * Bertl off to bed now ... have a good one everyone! 1288301976 N * Bertl Bertl_zZ 1288302047 M * ryker Bertl_zZ: same as the message it shows here: http://linux-vserver.org/Upstart_issues#Remove_console_outputs 1288302059 M * ryker except it's rsyslog and centos, not ubuntu using upstart 1288302070 M * ryker oh well, g'nite Bertl_zZ 1288303670 Q * michal Ping timeout: 480 seconds 1288303775 J * michal ~michal@www.rsbac.org 1288303791 Q * Hunger Ping timeout: 480 seconds 1288303895 Q * ntrs Ping timeout: 480 seconds 1288304342 J * ntrs ~ntrs@vault08.rosehosting.com 1288304613 Q * petzsch Quit: Leaving. 1288305573 Q * ghislain1 Quit: Leaving.