1105833609 M * SiD3WiNDR I have absolutely no clue about it :) 1105833617 M * SiD3WiNDR I can only test stuff and report bugs 1105833621 M * chrish01 nod 1105833631 M * SiD3WiNDR ... but I need hardware for it first, or at least vmware 1105833636 M * chrish01 i got decent understanding of kernel. but i wouldnt write any bugfree stuff =P 1105833646 M * chrish01 i use vmware to test his stuff right now 1105833652 M * chrish01 that way i can just do it on my dev box 1105833888 M * SiD3WiNDR :) 1105833899 M * SiD3WiNDR I understand stuff talked about but I cant find my way in kernel code 1105833904 M * SiD3WiNDR anyway, gotta get some sleep :) 1105833905 M * SiD3WiNDR 'night 1105833927 M * chrish01 night dude 1105834348 M * virtuoso Decent understanding is admirable, but I'd prefer qemu over vmware. 1105834407 M * chrish01 i have friends at vmware...so it works for me :) 1105834419 M * chrish01 i hear good things of qemu, but never used it myself 1105834487 M * chrish01 i must admin, qemu's arch support is much better :) 1105834518 M * virtuoso Friends at vmware? That's great. :) Care to talk them into providing you with sources? :) 1105834528 M * chrish01 haha fat chance 1105834537 M * chrish01 vmware 5 should be *very* smooth though 1105834542 M * chrish01 im trying to get on beta list 1105834550 M * chrish01 gtk2 gui in fact :) 1105834582 M * virtuoso Could be better with GPL instead of GTK. :) 1105834588 M * chrish01 nod 1105834600 M * virtuoso When's it scheduled? 1105834618 M * chrish01 not sure for release. prabably before april though 1105836507 J * serving ~serving@213.186.170.101 1105837450 Q * Rusty` Quit: 1105837597 Q * dsanta Remote host closed the connection 1105837599 J * dsanta ~santa@c68.190.156.105.roc.mn.charter.com 1105837615 Q * Pinnen Remote host closed the connection 1105837617 J * Pinnen ~pinnen@h194n2fls35o917.telia.com 1105838792 J * sannes ~ace@home.skarby.no 1105842114 Q * lilo Quit: 1105843501 Q * prae Quit: Ciao 1105849872 Q * chrish01 Remote host closed the connection 1105854461 J * Drone ~drone@c-67-168-71-150.client.comcast.net 1105855931 Q * Drone Quit: 1105857574 J * chrish01 ~chrish01@69.90.131.10 1105863243 Q * chrish01 Quit: chrish01 1105863471 J * chrish01 ~chrish01@69.90.131.10 1105865877 J * are|afk ~are@dsl-084-056-158-162.arcor-ip.net 1105865892 M * are|afk hi 1105865912 N * are|afk _are_ 1105867245 J * marmotte ~marmotte@ca-metz-4-50.w80-8.abo.wanadoo.fr 1105868358 Q * marmotte Quit: Leaving 1105868358 Q * sannes Read error: Connection reset by peer 1105868373 J * marmotte ~marmotte@ca-metz-4-50.w80-8.abo.wanadoo.fr 1105868429 Q * marmotte Quit: 1105869022 Q * berni_ jupiter.oftc.net plasma.oftc.net 1105869143 J * berni_ ~berni@2001:1b18:202::2 1105869461 N * Bertl_zZ Bertl_oO 1105869960 Q * Bertl_oO Ping timeout: 480 seconds 1105870263 J * marmotte ~marmotte@ca-metz-13-133.w80-8.abo.wanadoo.fr 1105871999 Q * marmotte Quit: Leaving 1105874852 Q * monrad Quit: Leaving 1105875199 J * sannes ~ace@home.skarby.no 1105875209 M * sannes :) 1105875211 M * sannes morning 1105875218 M * daniel_hozac morning. 1105876079 J * monrad ~monrad@213083190130.sonofon.dk 1105877599 J * nayco ~nayco@lns-vlq-47-nan-82-252-251-25.adsl.proxad.net 1105877609 M * nayco 'llo ! 1105878920 Q * ndim Ping timeout: 480 seconds 1105879152 J * ndim U2FsdGVkX1@helena.bawue.de 1105885305 M * _are_ hmpf, no luck with this opteron box, megaraid controller just claimed 2 broken disks, after a reboot it claims both are fine 1105885601 J * are|lunch ~are@dsl-084-056-158-237.arcor-ip.net 1105885601 Q * _are_ Read error: Connection reset by peer 1105885829 N * are|lunch _are_ 1105886700 Q * mjung Ping timeout: 480 seconds 1105887936 J * Bertl_oO ~herbert@janus.mc.tuwien.ac.at 1105887984 N * Bertl_oO Bertl 1105888076 M * Bertl morning folks! 1105888119 M * Loki|muh moin Bertl 1105888367 M * Bertl how'ya doing? 1105888385 M * _are_ hi Bertl 1105888402 M * Bertl hey _are_, everything fine? 1105888470 M * sannes :) 1105888478 M * Bertl morning sannes! 1105888497 M * sannes Bertl:) 1105888528 M * Bertl did you already adapt the bme to 2.6.11-rc*? 1105888556 M * Bertl (or is this planned?) 1105888631 J * crush ~looking@66-42-2-46.oak.dial.netzero.com 1105888639 M * Bertl welcome crush! 1105888665 M * crush hi folks, from california 1105888723 M * Bertl and the others? 1105888740 M * _are_ Bertl: yesterday the LSI megaraid claimed to have lost 2 disks, as i only noticed today I use the opportunity to migrate from reiserfs to xfs 1105888770 M * Bertl with 2.6.11-rc*? 1105888782 M * _are_ btw: after a reboot linux driver and raid bios claimed all to be fine. 1105888784 M * _are_ 2.6.10 1105888897 M * Bertl crush: ah, sorry missed the comma ;) 1105889480 M * sannes Bertl : havn't looked at it yet.. 1105889535 M * sannes I guess I could take a look at it now.. is there a lot of updates recently? 1105889550 A * sannes will let the source speak for itself.. heh 1105889576 M * Bertl there are a lot, but I don't think it affects bme too much 1105889584 M * Bertl s/it/they/ 1105889994 M * sannes first patch looks okay atleast .. 1105890206 M * _are_ if i mount some directory with mount --rbind into the filesystem of some vserver, then start the vserver it sees all. if I add another mount below the source directory, the running vserver won't notice it. this is because of ... 1105890206 M * _are_ the namespace and i understand it somewhat. But my vservers seem to forget all the mounts in the source (basically act as if it was --bind) at some not really specified time. 1105890217 M * _are_ amazingly, this is not true in all vservers the same time 1105890276 M * Bertl hmm ... 1105890382 M * _are_ i am not able to reproduce it, btw. first thought it has something to do with a mount operation in the main server, but i haven't done any. 1105890454 M * _are_ and from the mainserver context you can still see the full directories, as well as from some vservers (or at leats 1) and same time you can't see the submounts from at least 1 vserver 1105890544 M * Bertl basically it's a matter of what is mounted when ... 1105890562 M * _are_ well, everything is mounted, then the vservers are started 1105890569 M * Bertl --rbind mounts copy the entire mount tree recursively 1105890597 M * _are_ ok, so paranoid people better mount it one-by-one 1105890606 M * _are_ and I get really paranoid here. 1105890609 M * Bertl (within the current namespace) 1105890614 M * _are_ ofc. 1105890650 M * _are_ the namespace issue we had earlier. ;) 1105890650 M * Bertl if you create a new vserver, you get a new namespace (i.e. an --rbind copy of the current mounts _below_ the vserver root dir) 1105890698 M * Bertl I don't know if the tools call/activate the namespace cleanup functions ... (you can check that with the linux-vserver debug system) 1105890727 M * _are_ I am really missing some tool to display the mounts and the namespace/context they are visible in 1105890759 M * Bertl /proc/mounts should display that 1105890810 M * _are_ looks like fstab lines to me, or are the last 2*0 not the fstab values? 1105890822 M * _are_ (no vserver running atm as i migrate the data) 1105890924 M * Bertl /proc/mounts (if completely enabled) always shows the 'current' status of _all_ moutns inside the current namespace 1105891820 M * _are_ ah, ok, so i just enter each namespace and cat /proc/mounts. from main server it looks to me like I can see al mounts, this is still correct? 1105891846 J * mjung ~madbob@mpii00103.mpi-sb.mpg.de 1105892040 M * nayco Bertl: With the last util-vserver, I cannot start a vserver... http://pastebin.com/229540 1105892377 M * Bertl welcome mjung! 1105892449 M * Bertl nayco: hmm, which tools? 1105892779 Q * nayco Ping timeout: 480 seconds 1105893111 J * nayco ~nayco@lns-vlq-47-nan-82-252-254-101.adsl.proxad.net 1105893117 M * nayco back 1105893128 M * Bertl wb nayco 1105893135 M * Bertl < Bertl> nayco: hmm, which tools? 1105893176 M * nayco util-vserver-legacy-0.30.196-2mdk 1105893176 M * nayco util-vserver-devel-0.30.196-2mdk 1105893176 M * nayco util-vserver-0.30.196-2mdk 1105893176 M * nayco util-vserver-core-0.30.196-2mdk 1105893176 M * nayco util-vserver-build-0.30.196-2mdk 1105893177 M * nayco util-vserver-lib-0.30.196-2mdk 1105893179 M * nayco util-vserver-sysv-0.30.196-2mdk 1105893239 M * Bertl and they give you @MKFIFO@: command not found ? 1105893397 M * nayco Yes :D ! 1105893400 M * nayco :o ! 1105893422 M * nayco I work with other version of the tools... 1105893426 M * nayco +s 1105893431 M * Bertl hmm, strange ... those are normally replace by the autoconf part ... 1105893460 M * Bertl I guess we have to wait for enrico (or until I get some time to play with the tools) 1105893466 M * nayco I failed to compile them by myself, as I told yesterday, so I use the precompiled ones here. 1105893498 M * Bertl okay, dinnertime ... back in 30 or so ... 1105893498 M * nayco But do they work for you ? 1105893526 M * Bertl nayco, you have a configured/working system except for the tools, right? 1105893540 M * nayco You mean the host system ? 1105893548 M * Bertl is there a way to get an rpm-build-able account on your machine? 1105893570 M * Bertl (I'm still with some mandrake 8.2-9.1 version, and here everything works fine) 1105893577 M * nayco er... You want a user account, right ? 1105893595 M * nayco Well, This mchine is on ADSL, but we can try... 1105893602 M * Bertl something where I can rebuild the source rpm and check it ... 1105893621 M * nayco Well, wiwh rights must I give to the account ? 1105893633 M * nayco I put it in a given group ? 1105893678 M * Bertl try to figure out what is required (when you can recompile/rebuild an rpm it is fine for me ;) 1105893687 M * Bertl thanks, off now ... 1105893692 N * Bertl Bertl_oO 1105894322 M * daniel_hozac nothing special should be required. 1105894515 M * nayco You mean, no "rrot" rights ? 1105894520 M * nayco root 1105894551 M * nayco I can use a ~/.rpm*, no ? 1105894617 M * daniel_hozac right. 1105894639 M * daniel_hozac never build stuff as root ;) 1105894835 Q * mjung Ping timeout: 480 seconds 1105894839 M * nayco Ok, I'll check a rpm tutorial.... I saw one for Mandrake. 1105895207 P * crush 1105895254 J * mjung ~madbob@contact.mpi-sb.mpg.de 1105895318 N * Bertl_oO Bertl 1105895326 M * Bertl back 1105895904 M * nayco Bertl: what do you need to build without been root ? 1105895908 M * nayco being 1105895939 M * Bertl give me a few moments, maybe I've found a machine to test 1105895965 M * nayco k 1105898823 Q * mjung Remote host closed the connection 1105898839 J * mjung ~madbob@mpii00103.mpi-sb.mpg.de 1105898997 Q * sannes Read error: Connection reset by peer 1105899835 J * Pazzo ~thomas@host130-250.pool8172.interbusiness.it 1105899839 M * Pazzo hola! 1105899845 M * Bertl hey Pazzo! 1105899851 M * Pazzo hi bertl! 1105900019 M * Pazzo I'm trying to run different vservers using the host's main ip address - with the following little problem: stopping a vserver it takes down the hosts primary ip... is this my fault (config?) or didn't anyone think about this before? 1105900037 M * Bertl it's your fault! 1105900064 M * Pazzo thaught so :) 1105900071 M * Pazzo s/a/o/ 1105900078 J * miller7 ~none@adsl49-static-gw1.access.acn.gr 1105900087 M * Bertl welcome miller7! 1105900088 M * miller7 hello to everyone 1105900091 M * miller7 hey Bert 1105900096 M * Bertl quite some time ;) 1105900106 M * miller7 yep 1105900110 M * miller7 too busy 1105900126 M * miller7 how are you doing? Are you doing ok? 1105900161 M * Bertl I'm fine ... still busy with stuff I shouldn't do but .. hey so is life ;) 1105900168 M * miller7 :) 1105900175 M * miller7 I know what you mean 1105900181 M * miller7 I'm on same position 1105900358 M * Pazzo Bertl: Could you give me a little help? Flowerpage doesn't help me :-( wth am I doing wrong? 1105900401 M * Bertl you probably specify a device (like eth0) which is just _wrong_ if you want to _reuse_ any ip 1105900447 M * Pazzo I was testing this right now :) thnx bertl! 1105900458 M * miller7 bert, what are the recommended entries on /dev? 1105900484 M * Pazzo so if I create a "dev" file the ip will be removed, right? 1105900503 M * Bertl miller7: $ ls /dev/ 1105900503 M * Bertl full log= null ptmx pts/ random tty urandom zero 1105900518 M * Bertl (but recent tools know better ;) 1105900534 M * miller7 I'm creating the vserver on my own 1105900549 M * miller7 and I have all the entries there so I thought I'd delete some :D 1105900579 M * Pazzo works great :) 1105900663 M * Bertl Pazzo: well, the ip is created and removed, just the creation fails (because it's already there) 1105900718 M * Pazzo ?? the ip isn't removed without "dev"! 1105900779 M * Bertl without the dev, it means just _use_ this ip, nothing is created, and nothing removed ... 1105900872 M * Pazzo shouldn't this be noted on the flowerpage? documentation there is very confusing! 1105900900 M * miller7 what is the flowerpage? 1105900908 M * Bertl feel free to send patches/suggestions to enrico 1105900995 M * Pazzo with my good english? I'll let him know what I would try to explain there :) 1105901024 M * Bertl but basically this behaviour is there since the beginning ... 1105901064 M * Bertl http://vserver.13thfloor.at/Stuff/VServer-IP-Setup-0.1.txt 1105901108 M * SiD3WiNDR hm 1105901113 M * SiD3WiNDR in vserver-stat, what is the vsz line? 1105901118 M * SiD3WiNDR s/line/column/ 1105901139 M * Bertl the sum of all vm sizes 1105901280 M * Pazzo shit - I reinstalled our fileserver last week with vserver-kernel and I have been really happy that everything worked out-of-the-box. nfs, samba, dhcp - everything running well inside vserver. today I realized that the capability module wasn't loaded and now as it is loaded I was wondering why the nfs-server stopped working :-( 1105901355 M * daniel_hozac kernel nfsd just requires some capability? 1105901394 M * Pazzo hmmm... just a sec... I'm causing trouble with my interfaces... 1105901401 Q * serving Ping timeout: 480 seconds 1105901414 M * Bertl I don't think that kernel nfs server will _ever_ work inside a vserver properly ... 1105901465 M * daniel_hozac that's what i thought. 1105901471 M * Pazzo bertl: no more ip inside my vserver if I remove "interfaces/dev" 1105901478 M * Pazzo (ip address) 1105901499 M * Pazzo what about leaving "dev" there and create a "nodev" file? 1105901869 M * Bertl yeah, what about it? try it! ;) 1105902006 M * SiD3WiNDR Bertl: what does that mean exactly? :o It says "3GB" on 2 of them 1105902053 M * Bertl nayco: sorry, doesn't work on mdk > 9.x, please send a bug report to enrico, I tried to fix it, even with the autoconf/automake but it just doesn't work for me ... 1105902069 M * Bertl SiD3WiNDR: hmm? 1105902077 M * SiD3WiNDR the vsz 1105902098 M * Pazzo Bertl: just some secs - I'm creating some lvms remote somewhere else and don't want to kill my whole routing once again :) 1105902110 M * SiD3WiNDR I think it's rather odd that 2 say 3GB and the rest is 40-90MB 1105902117 M * Bertl SiD3WiNDR: it's the sum of all VM spaces assigned to the threads inside the vserver 1105902337 M * nayco Bertl: Ok, so the problem isn't my machine.... What is Enrico's site address ? 1105902423 M * Pazzo (back) 1105902523 M * Bertl nayco: hmm, seems I get at least 0.30.196 to work with a small trick ... could you try that on your machine? 1105902593 M * Bertl http://vserver.13thfloor.at/Stuff/MANDRAKE/util-vserver-0.30.196-3mdk.src.rpm 1105902630 M * nayco http://savannah.nongnu.org/bugs/?func=additem&group=util-vserver <= I found it ;-) 1105902637 M * nayco Ok, i'll try that ! 1105902641 M * nayco What is the trick ? 1105902659 M * Bertl it compiled here with: 1105902660 M * Bertl rpm -ba --nodeps /usr/src/RPM/SPECS/util-vserver.spec --without xalan --without doc 1105902661 M * Pazzo bertl: things don't work :( I'm using 0.30.196 - with "eth0" in interfaces/dev the primary ip will be removed if vserver stops - if I remove interfaces/dev stopping doesn't hurt but the vservre starts without ip address :( 1105902692 M * Bertl what does without ip address mean? 1105902701 M * Pazzo it gets no ip address 1105902727 M * Pazzo "ip address" inside the vserver shows nothing, services doesn't start 1105902733 M * Pazzo (don't) 1105902818 M * Pazzo interfaces/nodev seems to have no effect 1105902832 M * Bertl nayco: you might get the unpackaged message with that too, so try with a 'normal' rebuild first (and if it fails then do the following) 1105902871 M * Bertl rpm --rebuild util-vserver-0.30.196-3mdk.src.rpm >/tmp/util-vserver.log 2>/tmp/util-vserver.err 1105902885 M * Bertl and send the files via email ... 1105902916 M * nayco You want wich method first ? 1105902925 M * nayco I just started the first one... 1105902967 M * Bertl well, it doesn't matter ;) 1105903004 M * nayco Ok, I restarted the first method and I redirected the logs. 1105903042 M * Bertl Pazzo: please upload a representation of your # /etc/vservers//interfaces/ dir 1105903156 M * nox ho klar 1105903165 M * nox sorry 1105903166 M * Pazzo interfaces/0/ip => 192.168.124.1, interfaces/0/prefix => 24, interfaces/dev => eth0, interfaces/nodev 1105903183 M * Pazzo (192.168.124.1/24 is eth0's main ip address) 1105903212 M * Bertl well, remove the interfaces/dev 1105903296 M * Pazzo if I do that, stopping the vserver is fine - if I restart it there will be no ip available inside the vs 1105903324 M * Bertl please run the tools with --debug and upload the output somewhere 1105903397 M * nayco Bertl: What is the command line to set the locale, pls ? 1105903415 M * Bertl export LC_ALL=C LANG=C 1105903419 M * nayco thx 1105903491 M * Pazzo bertl: is it ok if I add another ip to the interface for doing the tests? otherwise I'm killing my fileserver/router all time :( 1105903502 M * miller7 I'm going home, take care guys 1105903508 M * Bertl cya 1105903515 P * miller7 1105903519 M * Bertl Pazzo: whatever you do it's fine for me ;) 1105903681 M * daniel_hozac Pazzo: you need interfaces/0/nodev. 1105903722 M * Bertl well, I assumed he had that ... because I asked for the dir 1105903743 M * Bertl but now I see he has dev and nodev one level higher? 1105903747 M * nayco Bertl: The building failed... you want the logs ? 1105903757 M * Bertl as I said ;) 1105903778 M * nayco ok. Where can I find your address ? 1105903786 M * Bertl herbert@13thfloor.at 1105903804 M * nayco Thx. In my address book ;-) 1105903873 J * Rusty Rusty@3e44a83a.adsl.enternet.hu 1105903875 N * Rusty Rusty` 1105903878 M * Rusty` hi guys 1105903883 M * Bertl welcome Rusty`! 1105903887 M * Pazzo hi rusty! 1105904212 M * nayco Bertl: Mail :) !!! 1105904231 M * Rusty` Bertl: i have a BME patch for linux kernel 2.6.10 :) 1105904254 M * Rusty` it is damned white of sannes to rediff for me... 1105904256 M * Pazzo bertl: http://vserver.13thfloor.at/Stuff/VServer-IP-Setup-0.1.txt 1105904289 M * Pazzo ehm ... sorry 1105904301 M * Pazzo it's this one: http://vserver.gelf.net/vserverstart-debug1.txt 1105905183 N * Pazzo PazZzzzzooo 1105906068 J * sannes ~ace@home.skarby.no 1105906551 Q * Rusty` Quit: 1105906557 J * Rusty Rusty@3e44a83a.adsl.enternet.hu 1105907153 M * Rusty there is a documentation about ngnet? 1105907312 J * pulsar ~pulsar@82.209.237.8 1105907412 M * pulsar What meaning of RSS in output vserver-stat ? Acording top its much more ... 1105907624 M * Bertl is this a new kind of fun sport? 1105907662 M * Bertl folks! vserver-stat is over one year old, and it is, was, has and will do always the same ... 1105907670 M * Rusty :) 1105907683 M * Bertl just sum up the values from ps auxww ... 1105907696 M * Bertl are they correct? no! are they relevant? no! 1105907704 M * pulsar hm 1105907721 M * Bertl do they giva an idea what actually happens? not at all! ;) 1105907745 M * nayco so this command is worth nothing ;) ? 1105907746 M * Bertl for appropriate accounting look in /proc/virtual//limit 1105907767 M * Bertl nayco: it's as much worth as ps auxwww ;) 1105907788 M * nayco :D ! 1105907803 M * nayco Was my mail useful ? 1105907832 M * Bertl well, yes and no ... 1105907859 M * Bertl I'm going to upload my latest packages, maybe they work for you too ;) 1105907860 M * nayco I France, we'd say that you are from normandy... ;-) 1105907868 M * nayco Ok. 1105907923 M * nayco So, the problems are specfile related ? 1105907965 M * Bertl the problems are util-vserver + mandrake related 1105908005 M * Bertl basically you are hitting 4 different intermangled issues 1105908012 M * nayco ...mmmmm 1105908021 M * Bertl a) files which are not created but expected to be there 1105908035 M * Bertl b) files which are created, installed but not packaged 1105908053 M * Bertl c) configure/autoconf/automake version issues 1105908074 M * Bertl d) incorrectly prepared helper files 1105908102 M * Bertl why don't I see that here? I don't know probably because my tools are outdated ... 1105908122 M * Bertl for example, my rpm doesn't complain about unpackaged files 1105908123 J * serving ~serving@213.186.180.245 1105908160 M * nayco well... 1105908166 M * Bertl nayco: basically rpm -tb util-vserver-0.30.196.tar.gz is supposed to work ... have you tried that? if yes, what was the result? 1105908177 M * nayco I try it. 1105908198 M * daniel_hozac Bertl: %define _unpackaged_files_terminate_build 0 ;) 1105908205 M * nayco er... Wich version of the tgz ? the one in .........3mdk ? 1105908218 M * Bertl take the one from enricos site ... 1105908225 M * nayco Ok. 1105908272 M * nayco http://www-user.tu-chemnitz.de/~ensc/util-vserver/util-vserver-0.30.tar.bz2 => 03-Jul-2004 ??? This one ? 1105908337 M * Bertl http://www-user.tu-chemnitz.de/~ensc/util-vserver/alpha/util-vserver-0.30.196.tar.bz2 1105908348 M * nayco k 1105908349 M * Bertl daniel_hozac: thanks for the info ... 1105908404 M * Rusty Bertl: there is documentation's for ngnet? 1105908422 M * Bertl nope, not really, just a script 1105908511 M * Rusty i see 1105908546 M * Bertl http://vserver.13thfloor.at/Experimental/NGNET/vnet3_setup.sh 1105908557 M * Bertl (basically does everything required to test/use it ;) 1105908730 M * Rusty what features have in ngnet? :) 1105908829 M * Bertl - virtual interfaces, routing, config 1105908847 M * Bertl - (later) virtual iptables, arp 1105908881 M * Bertl - support for ipv4, (later) ipv6, tun/tap, ppp ... 1105908941 M * Rusty mmh, it sounds good! 1105909256 M * pulsar I'm realy dumb. Anyone can tell me how can I get via vserver tools same values of RSS what I see in top in each vsp? 1105909358 M * Bertl where do you see rss values for more than one process in top? 1105909383 M * mugwump heh 1105909542 M * pulsar Seems like I'm totaly misunderstand meaning of RSS... 1105909580 M * Bertl maybe, what does it mean for you? 1105909587 M * Rusty Bertl: how soon may we expect the first "not for only kernel hacker's" version of ngnet? ;) 1105909654 M * pulsar Bertl, amount of physical ram alocated by process 1105909686 M * Bertl Rusty: as soon as some userspace hacker will hack for it ;) 1105909711 M * Bertl pulsar: processes do not allocate physical ram 1105909735 M * Rusty Bertl: i see :) 1105909819 M * pulsar Bertl: acordind my top manpage RES -- Resident size (kb) The non-swapped physical memory a task has used. 1105909837 M * Bertl yep, that's correct 1105909986 M * mugwump pulsar, the problem is - how do you count memory that is shared between processes, like shared objects (.so files) ? 1105909994 M * mugwump or even common binaries 1105910011 M * Bertl or threads? 1105910147 M * pulsar OK. For examle memory can not be shared betwin two different VPSs, can we calculate total amount of ram used by VPS ? 1105910185 M * Bertl in theory, yes, practically no 1105910209 M * Bertl (well at least not without slowing down everything) 1105910248 M * pulsar My boss want prety graphs with amount of ram used by each customer :-( 1105910285 M * Bertl hmm, what about file caches? 1105910295 M * Bertl what about the 'RAM' the kernel uses? 1105910315 M * Bertl network packet buffers? 1105910329 M * Bertl interrupt stacks for disk io? 1105910382 M * nayco Bertl: Do you want the log from the (failed) building of the alpha tools ? 1105910397 M * pulsar It's kernel related stuff, userspase RAM usage more informative for me 1105910399 M * Bertl nope, please send it directly to enrico or the mailing list 1105910408 M * nayco ok. 1105910418 M * Bertl pulsar: define userspace ram usage! 1105910495 M * Bertl nayco: btw, the 0.30.196-3mdk tools (compiled ones) work fine ehre, just tested ... 1105910506 Q * weasel Quit: reboot. bbib 1105910508 M * Bertl (maybe they work for you too) 1105910528 M * Bertl http://vserver.13thfloor.at/Stuff/MANDRAKE/ 1105910704 Q * Snow-Man Remote host closed the connection 1105910718 Q * nox uranium.oftc.net neutron.oftc.net 1105910735 M * pulsar Bertl, code+data ... I'm totaly wrong ? 1105910751 J * Snow-Man ~sfrost@snowman.net 1105910765 M * Bertl pulsar: you are wrong ... 1105910806 M * Bertl first, code pages are not loaded into memory unless referenced 1105910859 M * Bertl so if you have 1Mb of code, and just access the first 2Kb, then the rest will not even be loaded from disk 1105910881 M * Bertl then, data, really depends on what a program does ... for example 1105910905 M * Bertl java is well known, to reserve roughly 2GB of memory on startup ... 1105910946 M * Bertl does this mean you need 2GB memory to run java? no 1105910992 J * nox ~nox@c135225.adsl.hansenet.de 1105911044 M * Bertl pulsar: look at the values in /proc/virtual//limit 1105911051 M * pulsar Already 1105911077 M * pulsar RSS: 84565 137388 400000 0 1105911082 M * Bertl the VM and RSS will give you the number of pages allocated for each context 1105911090 M * pulsar yes 1105911113 M * chrish01 hey dudes 1105911123 M * Bertl hey dude! 1105911138 M * nayco 'loo ! 1105911159 M * chrish01 hi nayco 1105911182 M * pulsar Bertl, can You suggest me "realy good" reading about memory management ? 1105911198 M * Bertl hmm, let's see ... 1105911224 M * chrish01 Robert Loves kernel book is very good for kernel level stuff 1105911228 M * Bertl http://www.skynet.ie/~mel/projects/vm/ 1105911267 M * pulsar Thank You! 1105911300 M * Bertl http://www.cs.purdue.edu/homes/li/cs690Z/Outline/vmm.pdf 1105911311 M * Bertl http://home.earthlink.net/~jknapka/linux-mm/vmoutline.html 1105911333 M * Bertl http://www.tldp.org/HOWTO/KernelAnalysis-HOWTO-7.html 1105911471 M * Bertl nayco: do the compiled util-vserver packages work for you? 1105911525 Q * tchan Quit: leaving 1105911529 M * nayco ......3mdk ? The last ? 1105911541 M * Bertl yep (not the src rpm, the binaries) 1105911564 M * nayco nox: I can create a vserver, but can't start it. 1105911606 J * tchan ~tchan@c-24-13-81-164.client.comcast.net 1105911614 M * Bertl wb tchan! 1105911720 M * nayco Bertl: vserver test1 start 1105911720 M * nayco /usr/sbin/vserver: line 1: @MKTEMP@: command not found 1105911720 M * nayco /usr/sbin/vserver: line 163: @RM@: command not found 1105911720 M * nayco /usr/sbin/vserver: line 164: @MKFIFO@: command not found 1105911720 M * nayco Not enough parameters; use '--help' for more information 1105911723 M * nayco ( to quit :( ) 1105911770 M * Bertl are you sure you are using the -3mdk binaries? 1105911778 M * Bertl let's do the following: 1105911794 M * Bertl rpm -e util-vserver{,-lib,-core,-build,-sysv,-legacy} 1105911908 M * nayco Oh, no, I use the -2mdk series... Sorry. I'm downloading te last... 1105911955 M * mugwump does history tracing slow things down much? 1105911984 M * Bertl I have no stats for that ... feel free to do some testing/benchmarking ... 1105912026 M * Bertl it's designed to be lightweight (i.e. per cpu storage and such) 1105912054 M * mugwump cool, guess I'll leave it on then. I'm building 2.6.10+vs1.9.4-rc2 1105912173 M * nayco Bertl: Is the following error is important while installing util-vserver-0.30.196-3mdk.i586.rpm ? 1105912176 M * nayco /usr/bin/chattr: Operation not supported while setting flags on /vservers 1105912211 M * Bertl well, it sounds a little weird ... 1105912236 M * Bertl strange filesystem or kernel config? 1105912251 M * mugwump jfs /vservers? :) 1105912286 M * nayco I always got it. Anyway, I tried to start my test vserver: No message. But I cnnot enter it... I'm afraid I've got to build it again (Sniffle, that means 15 minuts downloading stuff...) 1105912346 M * Bertl nayco: here is what I did for a test: 1105912369 M * Bertl wget http://www.lycos-vds.com/dists/debian31-2004082701.tar.bz2 1105912412 M * Bertl vserver ZZZZ build -m skeleton --hostname ZZZZ.test.org --interface eth0:192.168.0.2/24 --context 101 1105912441 M * Bertl cd /vservers/ZZZZ; tar xjf /path/to/debian31-2004082701.tar.bz2 1105912453 M * nayco does this creates a eth alias ? 1105912493 M * Bertl no alias, if you want an alias, use: 1105912511 M * Bertl --interface ZZZZ=eth0:192.168.0.2/24 1105912669 Q * nox uranium.oftc.net neutron.oftc.net 1105912686 J * nox ~nox@c135225.adsl.hansenet.de 1105912818 M * Loki|muh Bertl: are there big important changes between 1.9.4rc1 and rc2? 1105912831 M * Loki|muh do you have a rc1 vs rc2? *g* 1105912836 M * Loki|muh for 2.6.89 1105912838 M * Loki|muh -9 1105912842 M * Loki|muh gna :( 1105912847 M * Bertl not if you do not use preemption ... 1105912850 M * Loki|muh for 2.6.9 1105912856 M * Loki|muh hmmm, I use it 1105912859 M * Loki|muh is that bad? 1105912879 M * Bertl the history stuff in rc1 isn't preemption safe ... 1105912917 M * Loki|muh history stuff? 1105912917 M * Bertl (i.e. it will blow up your machine sooner or later ;) 1105912921 M * Loki|muh ouch 1105912942 M * Bertl well, you probably do not have it enabled anyway ... 1105912962 M * Bertl (at least it's not on by default) 1105912985 M * Loki|muh CONFIG_PREEMPT=y 1105912987 M * Loki|muh ;) 1105913024 M * Bertl CONFIG_VSERVER_HISTORY ? 1105913068 M * Loki|muh Pattern not found 1105913068 Q * sannes Read error: Connection reset by peer 1105913087 M * Bertl so I guess it's off then ;) 1105913102 M * Loki|muh ah 1105913108 M * Loki|muh not I got it ;) 1105913118 M * Bertl s/not/now/ 1105913135 M * Loki|muh yeah 1105913155 M * Loki|muh the Weißbier is too good tonight *g* 1105913171 M * Loki|muh this history stuff is pure debugging? 1105913177 M * Bertl don't drink and admin/code ;) 1105913193 M * Loki|muh or are there another useful use-cases? 1105913219 M * Bertl you know those funny stack traces when your kernel kicks the bucket? 1105913230 M * Loki|muh uh, no 1105913243 M * Bertl well, lucky you then ;) 1105913413 M * Loki|muh all I work on tonight is non-production enviroment, so its save I think ;) 1105913775 M * nayco Bertl: With the debian image, this work.... Well, it seems ;-) 1105913806 M * Bertl good! ;) 1105913839 M * nayco Ok, i'll see tomorrow at work. Still got to make cups work... 1105913852 M * nayco and samba, and bind, and.... :p 1105913858 M * Loki|muh hehe 1105913862 M * Bertl ... and dream on! ;) 1105913906 M * nayco ..so hard to do :| ? 1105913951 M * Loki|muh samba and bind can be a long time 1105913956 M * Loki|muh depends on your setup 1105913974 M * Bertl and IIRC cups is doing bad things with 127.0.0.1 1105914053 M * Seraph re =) 1105914066 M * Seraph greetings from OpenTelephony Summit in Geilenkirchen 1105914077 M * Seraph Bertl: any progress on the 32 compat interface? ;) 1105914106 M * Loki|muh Geilenkirchen? I only know Gelsenkirchen *g* 1105914120 M * Seraph Geilenkirchen near Aachen 1105914128 M * Loki|muh ah 1105914130 M * Bertl Seraph: no, nothing new yet ... 1105914194 M * Seraph Bertl: ok ;) 1105914257 N * chrish01 chris|out 1105914889 M * nayco bye, all :) ! 1105914896 M * Bertl cya 1105914906 M * Rusty Bertl: what is the time at your place? 1105914919 M * Bertl 23:34 1105914920 P * nayco go to sleep ! 1105914939 M * Rusty hm, what is your place? :) 1105914974 M * Bertl hmm ... a room with a bunch of computers ;) 1105915091 M * Rusty i am lame in english.. 1105915158 M * Bertl universe, milky-way, sol-system, earth, europe, austria, pama (atm) 1105915185 M * Loki|muh pama is a city? 1105915230 M * Bertl well .. uhm .. it's more a village ... 1105915235 M * Rusty Bertl: u are nearly... :) 1105915239 M * Rusty i am from Hungary 1105915281 M * Rusty milky-way is a chocolate, no? :) 1105915419 M * Bertl yeah, choccolate bars have strange names, don't they? 1105915758 M * Bertl okay, folks, I'm off to bed now ... 1105915785 N * Bertl Bertl_zZ 1105916415 J * stephenM steve@user-133.l4a.c1.dsl.pol.co.uk 1105916420 M * stephenM hey all 1105916433 M * stephenM does anybody know if there's a patch for 2.6.10? 1105916516 M * Loki|muh http://vserver.13thfloor.at/Experimental/ 1105916530 M * stephenM cheers 1105916542 M * stephenM which file do I need :P 1105916547 M * stephenM for the most stable vserver system 1105916652 M * Loki|muh http://www.13thfloor.at/vserver/s_release/v1.29/ 1105916653 M * Loki|muh this 1105916664 M * Loki|muh but its against 2.4.x *g* 1105916675 M * stephenM what I mean is, I want the most stable version of vserver for 2.6.10 1105916711 M * Loki|muh i think this could be stable enough: http://vserver.13thfloor.at/Experimental/RC-1.9.4/ 1105916717 M * stephenM thanks 1105916806 M * Loki|muh but i think there are some problems with smp 1105916810 M * stephenM aha 1105916818 M * stephenM is it good enough to use in a production system? 1105916879 M * Loki|muh i use 1.9.3 actually on a few machines and will switch to 1.9.4rc1 soon 1105916897 M * stephenM is rc1 or rc2 the best would you say? 1105917032 M * Loki|muh rc2 is a bugfix which most of the user won't need as its only needed for preemption and vserver history 1105917111 M * stephenM ah 1105917129 M * stephenM do you know how to combine grsec and vserver? I'm trying but patch is reporting failures 1105917179 N * _are_ are|afk 1105917219 Q * pulsar Ping timeout: 480 seconds 1105917220 M * Loki|muh sorry, dunno 1105918744 J * Martin ~mjb@host-138-38-225-19.nat.bath.ac.uk 1105918762 M * Martin Hi 1105918794 M * Martin Sorry to trouble you all but had some questions about vserver and wondered if you could help. 1105918834 M * Martin Am looking to run an number of servers from a Sun E4000 using a virtualisation technology 1105918868 M * Martin Is vserver/sparc64 considered "production ready" 1105918916 M * Martin Also are there any known issues with OpenVPN (using tun devices) and / or NFS (need to run SFS which uses NFS loopback) 1105918930 M * Martin Any thoughts / experiences ? 1105919042 M * Loki|muh tunnels running independent are only possible with new network which is nuder development actually 1105919202 M * Martin Would it be possible to use tap or run the tunnels on the host system? 1105919275 P * albeiro irc is sooooooo boring 1105919415 M * Loki|muh sure, why not 1105919487 M * Loki|muh afaik, the vserver-patch has no impact on the host-system, it only allows to run vserver-guests 1105919528 M * Martin cool 1105919555 M * Martin If I get the system up and running I'll post an account fo how I did it. 1105919586 M * Martin Have to go work. Thanks 1105919589 Q * Martin Quit: using sirc version 2.211+KSIRC/1.1