1207095108 J * sladen paul@starsky.19inch.net 1207097672 Q * _gh_ Read error: Connection reset by peer 1207097695 J * _gh_ ~gerrit@c-67-169-199-103.hsd1.or.comcast.net 1207101421 Q * fatgoose Remote host closed the connection 1207101452 J * fatgoose ~samuel@76-10-149-199.dsl.teksavvy.com 1207102431 J * mick_work_ ~clamwin@h-74-2-196-226.miatflad.covad.net 1207102749 Q * mick_work Ping timeout: 480 seconds 1207102758 N * mick_work_ mick_work 1207102895 J * er ~sapan@pool-71-188-83-135.cmdnnj.east.verizon.net 1207103527 J * Infinito ~argos@201-14-70-159.gnace701.dsl.brasiltelecom.net.br 1207104027 Q * Infinito Quit: Leaving 1207104123 J * nauman ~nauman@pool-71-98-95-123.ipslin.dsl-w.verizon.net 1207105285 J * moeron ~moeron@pool-71-181-178-207.sctnpa.east.verizon.net 1207105863 Q * moeron Quit: Leaving 1207105936 Q * hparker Quit: Read error: 104 (Peer reset by connection) 1207107793 Q * nauman Ping timeout: 480 seconds 1207108111 J * balbir_ ~balbir@122.167.206.245 1207108538 Q * balbir Ping timeout: 480 seconds 1207109192 Q * dowdle Remote host closed the connection 1207109423 Q * balbir_ Read error: Operation timed out 1207110129 J * balbir_ ~balbir@122.167.218.252 1207110185 M * Bertl okay, off to bed now ... have a good one everyone! 1207110191 N * Bertl Bertl_zZ 1207113525 J * sharkjaw ~gab@64.28.12.166 1207113645 Q * balbir_ Read error: Operation timed out 1207114122 J * nauman ~nauman@pool-71-98-95-123.ipslin.dsl-w.verizon.net 1207114374 J * balbir_ ~balbir@122.167.179.114 1207114860 J * Slydder ~chuck@194.59.17.53 1207115653 J * virtuoso ~s0t0na@ppp91-122-102-135.pppoe.avangarddsl.ru 1207116063 Q * virtuoso_ Ping timeout: 480 seconds 1207118489 J * ftx ~ftx@dslb-084-062-248-244.pools.arcor-ip.net 1207118528 J * naren ~chatzilla@c-71-202-110-105.hsd1.ca.comcast.net 1207118536 Q * naren 1207118855 J * JonB ~NoSuchUse@77.75.164.169 1207119105 J * dna ~dna@88-203-dsl.kielnet.net 1207119373 Q * FireEgl Read error: No route to host 1207119479 J * dna_ ~dna@45-220-dsl.kielnet.net 1207119523 J * tobifix ~tobifix@IVV7KNALLER.UNI-MUENSTER.DE 1207119883 Q * dna Ping timeout: 480 seconds 1207120060 J * dna__ ~dna@204-224-dsl.kielnet.net 1207120468 Q * dna_ Ping timeout: 480 seconds 1207120977 Q * nauman Ping timeout: 480 seconds 1207121009 Q * JonB Quit: This computer has gone to sleep 1207121883 J * dna_ ~dna@217-236-dsl.kielnet.net 1207121981 J * dna ~dna@94-237-dsl.kielnet.net 1207122069 N * DoberMann[PullA] DoberMann 1207122287 Q * dna__ Ping timeout: 480 seconds 1207122302 J * dna__ ~dna@65-239-dsl.kielnet.net 1207122362 J * mrfree ~mrfree@host1-89-static.40-88-b.business.telecomitalia.it 1207122368 Q * dna_ Ping timeout: 480 seconds 1207122598 Q * dna Ping timeout: 480 seconds 1207122706 J * dna ~dna@35-243-dsl.kielnet.net 1207123058 Q * dna__ Ping timeout: 480 seconds 1207123647 J * yarihm ~yarihm@mtec-hg-docking-2-dhcp-130.ethz.ch 1207124221 N * ensc Guest288 1207124221 Q * Guest288 Read error: Connection reset by peer 1207124231 J * ensc ~irc-ensc@77.235.182.26 1207124328 J * anoncinfinity ~nwalsh@staffnet.internode.com.au 1207124346 Q * anoncinfinity 1207124484 J * dna_ ~dna@46-201-dsl.kielnet.net 1207124882 Q * dna Ping timeout: 480 seconds 1207125701 J * PowerKe ~tom@d54C13E4B.access.telenet.be 1207126636 Q * jsambrook Quit: Leaving. 1207126940 Q * balbir_ Remote host closed the connection 1207127712 J * JonB ~NoSuchUse@130.226.210.8 1207129518 J * arapaho ~arapaho@77.197.56.222 1207130657 Q * er Quit: er 1207130666 J * jsambrook ~jsambrook@anchor-internet-1-if0.router.demon.net 1207131244 J * friendly12345 ~friendly@ppp121-44-224-29.lns2.mel4.internode.on.net 1207131366 J * cryptronic ~oli@p54A3BD2C.dip0.t-ipconnect.de 1207131393 Q * JonB Quit: This computer has gone to sleep 1207131839 J * JonB ~NoSuchUse@130.226.210.8 1207131920 Q * JonB 1207132207 Q * mrfree Quit: Leaving 1207133974 J * balbir ~balbir@122.167.179.114 1207134972 Q * yarihm Quit: Leaving 1207135066 J * JonB ~NoSuchUse@130.226.210.8 1207135155 N * Bertl_zZ Bertl_oO 1207135825 Q * Aiken Remote host closed the connection 1207136223 Q * friendly12345 Quit: Leaving. 1207136262 J * yarihm ~yarihm@mtec-hg-docking-2-dhcp-130.ethz.ch 1207137667 M * ard6 vc_get_iattr(): Function not implemented 1207137670 M * ard6 Hmmmmm 1207137763 M * ard6 ard@tessa:~$ sudo vserver-stat 1207137763 M * ard6 can not change context: migrate kernel feature missing and 'compat' API disabled: Function not implemented 1207137793 M * ard6 there definetly is a compatability issue with 32bit's userspace and 64 bit kernel.... 1207137806 M * daniel_hozac sounds like you're not running a Linux-VServer kernel. 1207137809 A * ard6 reverts back to 2.6.22.19-vs2.3.0.34 1207137836 M * ard6 ard@tessa:~$ cat /proc/version 1207137836 M * ard6 Linux version 2.6.24.4-vs2.3.0.34-d64-xeon (root@etchdev64) (gcc version 4.1.2 20061115 (prerelease) (Debian 4.1.1-21)) #1 SMP Wed Apr 2 12:57:03 CEST 2008 1207137842 A * ard6 checks the config now 1207137879 M * ard6 ard@tessa:~$ grep VSERVER /boot/config-2.6.24.4-vs2.3.0.34-d64-xeon 1207137891 M * ard6 shows a CONFIG_VSERVER=y (and a bunch of others) 1207137909 M * daniel_hozac and CONIFG_IA32_COMPAT? 1207137910 M * ard6 and /proc/virtual exists... 1207137948 M * ard6 hmmm 1207137965 M * daniel_hozac maybe it's CONFIG_COMPAT. 1207137977 A * ard6 would assume that linux 32bits won't run if that is turned off.. 1207137980 M * ard6 CONFIG_COMPAT=y 1207137986 M * daniel_hozac true enough. 1207138019 M * daniel_hozac i'll add that to the list of 2.6.24 issues :) 1207138025 M * ard6 but ip6tables wasn't working either with 2.6.22.19 ... 1207138044 M * ard6 So, how do I upgrade the root environment to 64 bits... 1207138078 M * ard6 (and kqemu, and...) 1207138115 A * ard6 will do a network boot when he has time... 1207138150 M * ard6 Unless you want me to test the IA32 compat :-) 1207139385 Q * xdr Read error: Connection reset by peer 1207139503 Q * ftx Remote host closed the connection 1207140043 N * phedny Guest310 1207140048 J * phedny ~mark@2001:610:656::115 1207140053 Q * Guest310 Remote host closed the connection 1207141078 Q * JonB Quit: This computer has gone to sleep 1207141253 J * pmenier ~pme@LNeuilly-152-22-72-5.w193-251.abo.wanadoo.fr 1207141416 Q * sharkjaw Quit: Leaving 1207141740 J * JonB ~NoSuchUse@130.226.210.8 1207142429 Q * doener_ Ping timeout: 480 seconds 1207143679 Q * harry Quit: reboot, new grsec stuff 1207143733 N * Bertl_oO Bertl 1207143738 M * Bertl morning folks! 1207143835 J * harry ~harry@d51534A2F.access.telenet.be 1207144411 M * pmenier morning Bertl 1207144422 Q * ctrix Read error: Connection reset by peer 1207144425 J * dowdle ~dowdle@71-221-12-249.blng.qwest.net 1207144452 J * ctrix ~8__D@81-174-32-147.static.ngi.it 1207144708 J * insitu ~user@LMontsouris-152-62-50-113.w80-13.abo.wanadoo.fr 1207144713 M * insitu Hello, 1207144731 M * JonB hi 1207144750 M * insitu I am trying to install a new vserver using 'clone' method which used to work some months ago but is no more recognized on my new system 1207144765 M * JonB which util-vserver version? 1207144785 M * insitu I am on 0.30.212, debian etch 1207144803 M * JonB i think there is a 215 1207144858 M * insitu is there some debian package for it ? 1207144897 M * JonB maybe in backports 1207144917 M * Bertl insitu: but how did it vanish? 1207144926 Q * JonB Quit: This computer has gone to sleep 1207144948 M * Bertl insitu: ah, new system ... missed that part 1207144952 M * insitu server crashed !! (or rather was partially hacked, shame on me) 1207144970 M * Bertl you should be able to get at least 0.30.214 from backports 1207144982 M * Bertl and yeah, debian policy sucks :) 1207144998 M * insitu I kept a log but did not trace versions so it seems i used some newer features 1207145004 M * insitu ok, i look into backports 1207145039 M * insitu according to packages.debian.org, testing -> 0.30.214-6 1207145209 J * xdr ~xdr@19-173-96-87.cust.blixtvik.se 1207145285 M * Bertl wb xdr! 1207145345 M * insitu works like a charm! thanks a lot 1207145345 M * xdr hello Bertl :) omg ...buildworld on a 266mhz isn't something you should wait on 1207145372 M * Bertl insitu: you're welcome! feel free to hang around! 1207145375 M * Bertl xdr: hehe 1207145528 M * insitu hmmm. I am back quite quickly. In my logs, there was the following command 1207145531 M * insitu find /home/vserver/ -type f -exec /usr/sbin/setattr --iunlink {} \; 1207145561 M * Bertl not a good idea per se :) 1207145592 M * Bertl what you probably want to do is hashify/unify (via vserver) 1207145596 M * insitu sure: this makes /etc/mtab unwritable and lot of other files. 1207145614 M * insitu I think I mixed different things: cloning and hashifying no ? 1207145632 M * Bertl probably but even for hashification, this is the wrong way to do it 1207145651 M * insitu ok. could you point me to the right way (tm) :) ? 1207145663 M * insitu please 1207145726 M * Bertl http://linux-vserver.org/Frequently_Asked_Questions#What_is_vhashify.3F 1207145755 M * Bertl you have a kind of mini-howto right below that ... 1207145781 M * insitu yes, seen that already. did it change recently ? 1207145799 M * Bertl nope, not really 1207145817 M * insitu ok, just fiddling with things I barely understand then... 1207145829 Q * yarihm Quit: This computer has gone to sleep 1207145936 J * dna__ ~dna@112-216-dsl.kielnet.net 1207146022 M * Bertl insitu: if you have specific questions .. do not hesitate to ask 1207146130 Q * tobifix Quit: Leaving 1207146164 M * insitu yes. thanks again for your help. this is a really noob-friendly channel. not always the case elsewhere. 1207146166 M * insitu bye. 1207146168 P * insitu ERC Version 5.2 (IRC client for Emacs) 1207146343 Q * dna_ Ping timeout: 480 seconds 1207146450 Q * Wonka Ping timeout: 480 seconds 1207146680 J * hparker ~hparker@linux.homershut.net 1207147294 J * Hoochster ~hooch@xdsl-69-55-143-83.sofnet.net 1207147590 Q * hparker Remote host closed the connection 1207147724 J * hparker ~hparker@linux.homershut.net 1207148036 J * dna_ ~dna@38-195-dsl.kielnet.net 1207148442 Q * dna__ Ping timeout: 480 seconds 1207148837 M * Hoochster Would it be possible for someone to explain to me how to create the dev nodes I need to run an xserver inside a vserver, I have never had to do that before and am kind of stumped. thanks 1207148883 Q * balbir Remote host closed the connection 1207148921 M * PowerKe Hoochster: Create them from the host 1207148947 M * Hoochster sorry for the ignorance but how do I do that, like I said, I have never created any extra devices before so not quite sure of the syntax 1207148982 M * PowerKe mknod NAME TYPE [MAJOR MINOR] 1207149007 M * Hoochster so do I do this in the Vservers config directory? 1207149044 M * PowerKe You would create them in the guest directory, eg: mknod /vservers/myguest/dev/null c 1 3 1207149099 M * PowerKe you can even use cp -a /dev/null /vservers/myguest/dev/null to create it 1207149127 Q * Slydder Quit: Leaving. 1207149289 M * Hoochster I think the cp -a part was easiest to understand, thanks a ton, I think that worked 1207149570 M * Hoochster now my next dumb question heh, I did this for tty7, I am running an xserver on the host, and I notice the wiki mention use tty8 in the vserver, I am trying to use KDM for my login manager to autologin, do you know where I define to tell the vserver to startup on tty8 instead of 7? 1207149648 J * BobR odie@IRC.13thfloor.at 1207149649 J * xipe ~xipe@art95-1-82-232-246-225.fbx.proxad.net 1207149660 M * xipe Hi 1207149710 M * xipe I have a problem using traceroute from inside a vserver, I added the NET_RAW bcap and RAW_ICMP ccap 1207149714 M * xipe but I get 1207149717 M * PowerKe Hoochster: maybe you could try with openvt -c 8 ? 1207149721 M * xipe traceroute 195.242.99.126 1207149721 M * xipe traceroute to 195.242.99.126 (195.242.99.126), 30 hops max, 40 byte packets 1207149723 M * xipe traceroute: sendto: Operation not permitted 1 traceroute: wrote 195.242.99.126 40 chars, ret=-1 1207149741 M * xipe What have I done wrong ? :) 1207149761 M * Bertl xipe: nothing 1207149774 M * Bertl traceroute is a non-ip utility, which requires raw sockets 1207149789 M * Bertl it is better to use tracepath, which doesn't have this limitation 1207149801 M * PowerKe Hoochster: as openvt -sc 8 startx (not sure if it'll work) 1207149806 M * xipe Aren't raw socket provided by NET_RAW ? 1207149860 M * Bertl yes, did you give that to your guest? 1207149894 J * nauman ~nauman@pool-71-98-95-123.ipslin.dsl-w.verizon.net 1207149907 M * xipe yes 1207149921 M * xipe I added NET_RAW in the bacapabilities file 1207149928 M * xipe bcapabilities 1207149964 M * xipe before that I had a different error message 1207150025 M * xipe After adding the NET_RAW cap, the error message changed to the one I copied below 1207150042 M * xipe and now I am locked :) 1207150062 M * Bertl means you did shoot yourself in the foot, yes? 1207150089 M * xipe No, only that I don't understand the error message now :) 1207150154 M * Bertl it probably takes the device into permisc mode too 1207150165 M * Bertl which requires NET_ADMIN 1207150177 M * Bertl but what is the purpose of this anyway? 1207150234 M * xipe It's because I have 1 vserv over 3 that can't ping a certain host, but can tracepath without problem 1207150246 M * xipe I want to know what is not working 1207150260 J * tobifix ~tobifix@muedsl-82-207-197-141.citykom.de 1207150266 M * xipe If it's my configuration or my network provider that's having problem 1207150277 M * xipe that why I need traceroute :) 1207150290 M * Hoochster thanks for your help PowerKe I still got issues but I will keep working on them, both X servers seem to be starting but they don't seem seperated so will have to keep researching but you def helped me get started 1207150295 M * tobifix huhu 1207150296 M * tobifix ;) 1207150302 M * Bertl wb tobifix! 1207150309 M * tobifix hey Bertl 1207150319 M * tobifix already saw you making some traffic, nice ;) 1207150333 M * Bertl xipe: but why on the guest? 1207150349 M * xipe it works from the master ... 1207150357 M * xipe only this guest is not working 1207150363 M * Bertl tobifix: yeah, testing a little ... 1207150453 M * Bertl tobifix: btw, quite nice machine, and you forgot to send me the Hall'o'Fame data .) 1207150487 M * tobifix args, yes, i'll try to get that on my todo for this evening 1207150530 M * Bertl ah, no problem with that ... :) 1207150535 Q * xdr Read error: Connection reset by peer 1207150537 M * tobifix hehe 1207150674 M * Bertl xipe: yeah, but why not trace from the host with the guest ip? 1207150732 M * Bertl (i.e. use the -i and -s options) 1207150754 J * xdr ~xdr@19-173-96-87.cust.blixtvik.se 1207150777 M * Bertl xipe: what's your routing/host setup? 1207150842 M * xipe Just tried it works from the master host 1207150856 M * xipe So it only dosn't works from the vserver 1207150866 M * xipe The setup is like that : 1207150875 M * xipe I have 3 vservers 1207150888 M * xipe with 1 ppublic IP address each 1207150898 M * xipe and a master host with it's own IP address 1207150912 M * Bertl okay, so you have several 'public' ips assigned from your upstream provider, yes? 1207150918 M * xipe yes 1207150935 M * Bertl okay, let's try the following simple test: 1207150993 M * Bertl 'ping -c 5 -I www.google.com 1207151009 M * Bertl replace with all the public ips one after the other 1207151082 J * DavidS ~david@85.125.165.34 1207151087 M * Bertl wb DavidS! 1207151138 M * xipe It doesn't work with the ip of the vserver that's not working 1207151140 M * DavidS hi bertl! -> http://www.debienna.at/Kalender/2008-04-03 1207151145 M * DavidS (tomorrow) 1207151156 M * Bertl xipe: see, so that's an upstream issue then (or your routing is wrong) 1207151205 M * xipe I will open a ticket for my provider then, thank you very much :) 1207151208 M * PowerKe xipe: You're sure you're not using your network/broadcast ip number (just a thought)? 1207151223 M * xipe No I am not 1207151226 N * BobR BobR_oO 1207151240 M * Bertl xipe: unfortunately, that is a common mistake :) 1207151250 M * xipe :) 1207151597 Q * tobifix Remote host closed the connection 1207151738 J * tobifix ~tobifix@muedsl-82-207-197-141.citykom.de 1207151848 Q * nauman Ping timeout: 480 seconds 1207152431 J * Wonka produziert@chaos.in-kiel.de 1207152492 J * nicki_ ~nicki@sphinx.link-m.de 1207152504 M * Bertl welcome nicki_! 1207152516 M * nicki_ Hi, I'm trying to raise the kernel.shmmax value in one of my vserver guests. 1207152566 M * nicki_ I already created the file "/etc/vservers/spam5/sysctl/0/setting" with "kernel.shmmax" and "/etc/vservers/spam5/sysctl/0/value" with "134217728" 1207152588 M * Bertl what kernel and util-vserver version? 1207152608 M * nicki_ But when I start the guest, this parameter is still at:root@spam5:/# sysctl -a|grep shm 1207152608 M * nicki_ kernel.shmmax = 33554432 1207152632 M * nicki_ I'm running 2.6.22.18 with vserver patch 2.2.0.6 1207152693 M * Bertl and util-vserver is? 1207152717 M * nicki_ util-vserver (0.30.212-1) 1207152728 M * nicki_ System ist Debian Etch 1207152734 M * Bertl quite old, try with 0.30.214 or 0.30.215 1207152760 M * nicki_ Bertl: Do you know of any Debian packages for these new versions? 1207152767 M * Bertl you can get at least 0.30.214 from backports 1207152940 M * nicki_ Bertl: Darn. This worked. Thanks. 1207152945 M * Bertl you're welcome! 1207152959 M * Bertl ah, and please report it as bug to the debian folks :) 1207152976 M * nicki_ Bertl: I will 1207153023 M * DavidS who will probably close it without action, because it isn't release critical and fixed in lenny 1207153126 J * balbir ~balbir@122.167.179.114 1207153241 J * nauman nauman@pal-179-018.itap.purdue.edu 1207153628 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d 1207155212 Q * pmenier Quit: Konversation terminated! 1207155673 J * ftx ~ftx@dslb-084-062-252-125.pools.arcor-ip.net 1207155773 N * DoberMann DoberMann[PullA] 1207156213 J * JonB ~NoSuchUse@77.75.164.169 1207156396 Q * nicki_ Quit: Leaving 1207157226 Q * balbir Read error: Operation timed out 1207157934 J * balbir ~balbir@122.167.178.28 1207159025 J * virtuoso_ ~s0t0na@ppp91-122-170-17.pppoe.avangarddsl.ru 1207159027 Q * virtuoso Read error: Connection reset by peer 1207159357 Q * jsambrook Quit: Leaving. 1207159660 Q * ftx Remote host closed the connection 1207159685 J * ftx ~ftx@dslb-084-062-252-125.pools.arcor-ip.net 1207160782 Q * hparker Quit: Quit 1207160931 Q * JonB Quit: This computer has gone to sleep 1207160943 J * Piet ~piet@tor.noreply.org 1207161062 Q * xdr Ping timeout: 480 seconds 1207161647 J * dowdle_ ~dowdle@67-42-168-210.blng.qwest.net 1207161733 J * hparker ~hparker@linux.homershut.net 1207161878 Q * dowdle Ping timeout: 480 seconds 1207162424 J * JonB ~NoSuchUse@77.75.164.169 1207162790 N * dowdle_ dowdle 1207163437 Q * JonB Quit: This computer has gone to sleep 1207163898 Q * dna_ Ping timeout: 480 seconds 1207164345 Q * tobifix Remote host closed the connection 1207164466 J * dennis ~dennis@dslb-084-059-110-109.pools.arcor-ip.net 1207164483 M * Bertl wb dennis! 1207164493 M * dennis hi Bertl :-) 1207164610 J * JonB ~NoSuchUse@77.75.164.169 1207164667 M * dennis know you any script that manipulate config file like this: confedit /etc/conf.d/hostname hostname newhostname (file key value) ... so in config file the line HOSTNAME="oldhostname" will replace with HOSTNAME="newhostname" 1207164748 Q * nauman Ping timeout: 480 seconds 1207164780 M * ard dennis : sed -i 1207164803 M * Bertl dennis: sed? 1207164826 M * ard sed -i 's/^HOSTNAME=.*$/HOSTNAME="newhostname"/g' 1207164844 M * ard adjust regex to your liking 1207164949 M * dennis ahh... okay. thanks. :-) 1207164955 M * dennis i'll play with it. :) 1207165011 M * ard Be aware however that you should check for disk-full exceptions :-), although sed -i will probably do that 1207165053 J * larsivi ~larsivi@144.84-48-50.nextgentel.com 1207165055 M * ard (some people are able to bring down a whole supermarket chain with those minor bugs) 1207165112 A * ard off to bed 1207165134 M * Bertl ard: have a good sleep 1207165143 M * ard tnx! 1207165145 M * dennis okay 1207165150 M * dennis ard: gn8. :) 1207165971 J * yarihm ~yarihm@84-75-103-252.dclient.hispeed.ch 1207166416 J * doener ~doener@i577AFAF8.versanet.de 1207167085 M * dennis mh... why returns egrep '^HOSTNAME="(.*)"$' ./hostname the complete line and not only the value in braces? 1207167110 M * DavidS egrep is not sed 1207167123 M * dennis yes, its for reading a value. ;) 1207167158 M * dennis i think, thats egrep the better tool as sed 1207167267 M * daniel_hozac x=$(. ./hostname; echo $HOSTNAME) 1207167442 M * cehteh http://www.froscon.de/en/program/call-for-papers.html << notify! 1207167461 M * dennis its not only for hostname. its to read any value from a config file. 1207167555 M * daniel_hozac so? replace variable and filename as appropriate. 1207167599 Q * DavidS Quit: Leaving. 1207167663 M * dennis ahh... now i understand the script. :) 1207167678 M * dennis sorry, i'm confused ;) 1207168337 M * dennis daniel_hozac: its possible, thats $HOSTNAME is a var, like $1? (export $1=HOSTNAME) 1207168364 M * daniel_hozac huh? 1207168453 Q * Piet Quit: Piet 1207168458 M * cehteh .. :) 1207168478 Q * fatgoose resistance.oftc.net osmotic.oftc.net 1207168478 Q * nkukard resistance.oftc.net osmotic.oftc.net 1207168478 Q * nenolod resistance.oftc.net osmotic.oftc.net 1207168478 Q * besonen_mobile_ resistance.oftc.net osmotic.oftc.net 1207168478 Q * brc resistance.oftc.net osmotic.oftc.net 1207168481 M * Bertl that line actually works, but it is quite unusual :) 1207168493 M * dennis yes, it works... 1207168531 M * daniel_hozac using ${$1} rather than HOSTNAME looks rather ugly though, IMHO ;) 1207168535 M * dennis but the problem is, that i'll create a script thats forward the parameters (filename and value to get) to the command from daniel_hozac 1207168554 M * dennis config-get /etc/conf.d/hostname hostname 1207168557 M * daniel_hozac i gotta ask... how does this relate to Linux-VServer? 1207168568 M * dennis ähm, nothing. ;-) 1207168608 M * dennis its a little bit off topic. ;) 1207168734 J * fatgoose ~samuel@76-10-149-199.dsl.teksavvy.com 1207168734 J * nkukard ~nkukard@196.212.73.74 1207168734 J * nenolod ~nenolod@ip70-189-74-69.ok.ok.cox.net 1207168734 J * besonen_mobile_ ~besonen_m@71-220-235-129.eugn.qwest.net 1207168734 J * brc bruce@megarapido.cliquerapido.com.br 1207169470 Q * ftx Remote host closed the connection 1207169698 Q * bonbons Quit: Leaving 1207169974 J * hparker_lappie ~hparker@linux.homershut.net 1207170472 N * virtuoso_ virtuoso 1207170963 Q * mick_work Remote host closed the connection 1207171206 J * Hooch ~hooch@74.194.108.54 1207171264 Q * Hoochster Quit: Leaving 1207171276 J * Aiken ~james@ppp121-45-192-61.lns1.bne1.internode.on.net 1207171288 N * Hooch Hoochster 1207171653 Q * JonB Quit: This computer has gone to sleep 1207172075 J * dna_ ~dna@38-195-dsl.kielnet.net 1207172102 J * nauman ~nauman@pool-71-98-95-123.ipslin.dsl-w.verizon.net 1207172490 Q * cryptronic Quit: Leaving. 1207172503 Q * yarihm Quit: Leaving 1207173525 J * xdr ~xdr@172-173-96-87.cust.blixtvik.se 1207173651 Q * dna_ Quit: Verlassend 1207174041 J * lownoize ~lownoize@p5B076D65.dip.t-dialin.net 1207174041 Q * duckx Read error: Connection reset by peer 1207174044 M * lownoize hi 1207174058 J * duckx ~Duck@81.57.39.234 1207174210 M * lownoize i have a strange problem after a power failure i can not start one vserver anymore because it tells me that is already runing, all other vserver on this host work out of the box like before 1207174289 M * lownoize ah found the error, there was still a file under /var/run/vservers 1207174308 Q * larsivi Ping timeout: 480 seconds 1207174512 M * Bertl lownoize: okay, so problem solved? 1207174577 M * lownoize yes 1207174628 M * lownoize was just a bit confused because i got errors like vnamespace: vc_xidopt2xid("/etc/vservers/internal"): No such file or directory when i treid to restart/enter the vserver 1207175033 Q * nauman Ping timeout: 480 seconds 1207175532 J * jazzanova ~boris@S010600146cfc7d5b.vc.shawcable.net 1207175534 M * jazzanova hi 1207175546 M * jazzanova i'm trying to setup my vserver so it has no swap 1207175566 M * Bertl you mean the host 1207175574 M * jazzanova no 1207175579 M * jazzanova th eguest 1207175594 M * Bertl well, you can't simply make the guest _have_ no swap 1207175609 M * Bertl but, you can make the guest 'show' no swap 1207175683 M * jazzanova i'd like my program to crash if tries to allocate more memory than it is supposed to 1207175729 M * Bertl hmm, you have to turn overcommitment off for that, and set a strict RSS limit 1207175797 M * jazzanova 40004 19 103.5M 45.1M 1m46s16 0m00s87 7m49s88 francisco 1207175851 J * JonB ~NoSuchUse@77.75.164.169 1207175906 M * jazzanova boris@dagny:/etc/vservers/francisco/rlimits$ cat rss.hard 1207175906 M * jazzanova 128000 1207175907 M * jazzanova boris@dagny:/etc/vservers/francisco/rlimits$ cat as.hard 1207175907 M * jazzanova 128000 1207175923 M * jazzanova where's this overcommitment setting ? 1207176015 M * daniel_hozac /proc/sys/vm/overcommit_memory 1207176034 M * jazzanova this is on the host ? 1207176043 M * jazzanova or guest 1207176046 M * Bertl host 1207176056 M * jazzanova would it disable swap on the host too ? 1207176132 M * Bertl doesn't disable swap at all 1207176137 M * jazzanova this file has 0 now 1207176148 M * Bertl just makes the kernel check memory strictly 1207176196 M * Bertl 0: heuristic overcommit (this is the default) 1207176196 M * Bertl 1: always overcommit, never check 1207176197 M * Bertl 2: always check, never overcommit 1207176408 M * jazzanova and which one is for me ? 1207176419 M * jazzanova 2 ? 1207176429 M * Bertl well, you need 2, if you want to catch any program at the point of allocation 1207176462 M * Bertl note that linux heavily relies on overcommitment 1207176511 M * jazzanova what does the rss.hard limit do if it doesn't stop the VM from consuming memory ? 1207176531 M * Bertl the RSS is the Resident Memory Set ... 1207176543 M * Bertl i.e., what's actually instantiated in RAM 1207176557 M * Bertl compared to the amount _allocated_ by the program 1207176565 Q * JonB Quit: This computer has gone to sleep 1207176688 M * jazzanova so am I risking to get the host in trouble by setting it to 2 ? 1207176705 M * Bertl definitely 1207176747 M * jazzanova is it safer just turning off swap on the host ? 1207176775 M * daniel_hozac you pretty much never want to do that. Linux works best with some swap, even if it's just a couple of MiB. 1207176775 M * Bertl no, doesn't help you with this (the allocation) 1207176913 M * jazzanova i understood that AS is RAM + swap, so if I set AS and RSS to the same tihng, that means no swap 1207176956 M * Bertl nah 1207176962 M * Bertl you got that wrong 1207176980 M * Bertl AS is a something which really just makes sense for a single task 1207176988 M * Bertl (the address space) 1207177008 M * jazzanova ok 1207177016 M * Bertl in Linux-VServer, to give folks an idea about the general usage, we simply create a metric (by summing up the AS over all tasks) 1207177049 M * Bertl this is an artificial value and limit, nevertheless, it will act at allocation time 1207177063 M * Bertl RSS OTOH is a very _real_ value _and_ limit 1207177076 M * jazzanova so RSS overrides AS 1207177091 M * Bertl no, they are (almost) orthogonal 1207177110 M * Bertl they hold the relation RSS << AS 1207177146 M * jazzanova ok, so if i have two processes consuming 50mb of RAM, than AS = 100 ? 1207177156 M * Bertl the problem with RSS is, you do not know at allocation time, except in the case where you disable overcommitment 1207177188 M * Bertl so, on a typical setup, with overcommitment enabled, the allocation just 'consumes' AS, not RSS 1207177216 M * Bertl once that address space is actually used, the OOM killer will be invoked when the guest is over limit 1207177261 M * jazzanova ok 1207177315 M * jazzanova so if I want to limit my vserver processes to never use more than 400mb or memory, be it swap or ram, do I set AS to 400mb ? 1207177343 M * Bertl nope, you set RSS.hard to that 1207177363 M * jazzanova whath should I set AS too ? 1207177364 M * Bertl (assuming a recent stable kernel, that is) 1207177366 M * jazzanova doesn't matter ? 1207177373 M * jazzanova yeah, its recente. 1207177376 M * Bertl doesn't really matter for this purpose 1207177414 M * jazzanova so, now if on top of that I'd like to swap to be used, what does it have to do with overcommitment ? 1207177434 M * jazzanova from what I understand now, i have to turn off all swap on the host. 1207177447 M * jazzanova i'd liko no swap to be used 1207177471 M * Bertl if you want to avoid that the host uses swap, you have to turn off swap 1207177505 M * jazzanova i want to avoid that the guest uses swap. theer are no memory bugs in whath host runs, only in what guest runs. 1207177531 M * Bertl there is no way to do that selectively per guest atm 1207177543 M * Bertl memory management is done on the host 1207177565 M * jazzanova ok 1207177572 M * jazzanova i have 2gb of ram on the host 1207177597 M * jazzanova and suppose i never use more ram for host processes than 1600mbs 1207177610 M * jazzanova which leaves 400mbs for the guest 1207177627 M * jazzanova in this case, will guest ever use swap even if swap is enabled ? 1207177642 Q * xdr Remote host closed the connection 1207177670 M * Bertl if the guest is limited to < (400M - typical caches) 1207177696 M * Bertl then the host will not put memory on swap without being told to do so (swap policy) 1207177712 J * xdr ~xdr@172-173-96-87.cust.blixtvik.se 1207177714 M * jazzanova where is the swap policy setup ? 1207177860 M * Bertl check kernel/Documentation for current details 1207177870 M * Bertl (changes every second version) 1207178622 M * jazzanova ok, i have VSZ 430MB, and RSS 93RSS -- what does that mean ? 1207178631 M * jazzanova thans from vserver-stat 1207178795 M * Bertl what does /proc/virtual//limits say about AS and RSS? 1207178843 M * xdr I really hate that compaqs bioses sucks so hard 1207178871 M * xdr I can't find out anything useful from them 1207178903 M * jazzanova i got no AS there 1207178913 M * jazzanova RSS: 22942 0/ 28484 128000/ 128000 0 1207178914 M * Bertl it's called VM 1207178923 M * jazzanova VM: 110272 0/ 114739 128000/ 128000 0 1207178944 M * Bertl so that means that your guest currently holds 22942 pages of RAM 1207178963 M * Bertl and it allocated 110272 pages total (address space, that is) 1207178986 M * jazzanova that means that it uses 110272-22942 swap ? 1207179007 M * Bertl nope, not at all 1207179009 M * xdr jazzanova: ram not swap 1207179019 M * Bertl the 110272-22942 is 'unused' 1207179023 M * xdr and doesn't it mean reserved? 1207179039 M * Bertl nope, it is not reserved either, it is overcommited :) 1207179181 M * jazzanova hows the 22942 pages are different from 110292-22942 pages ? 1207179203 M * jazzanova are both sets loaded into ram ? 1207179277 M * Bertl no, only the 22942 pages are in RAM 1207179288 M * Bertl the others probably do not even exist 1207179292 M * jazzanova ok 1207179327 M * jazzanova so the programs allocated memory but never actually wrote into it ? 1207179329 M * Bertl (but could also be file backed or swap) 1207179347 M * Bertl yes, that's it most of the time 1207179362 M * jazzanova and file-backed means its mmaped ? 1207179377 M * Bertl take java for example ... it allocates 2GB when started, regardless of the actual memory used 1207179389 M * jazzanova ): 1207179416 M * Bertl okay, kind of sleepy here .. so I'm off to bed now 1207179424 M * Bertl have a good one everyone and cya! 1207179428 M * jazzanova thanks Bertl. 1207179430 N * Bertl Bertl_zZ 1207179623 Q * xdr Read error: Connection reset by peer 1207180293 J * nauman ~nauman@pool-71-98-95-123.ipslin.dsl-w.verizon.net 1207180463 J * Infinito ~argos@201-3-21-127.gnace701.dsl.brasiltelecom.net.br