1186099700 J * linus ~linus@bl7-129-104.dsl.telepac.pt 1186100991 Q * FireEgl Quit: Bye... 1186102248 Q * Piet_ Quit: Piet_ 1186102311 J * DoberMann_ ~james@AToulouse-156-1-180-178.w90-38.abo.wanadoo.fr 1186102419 Q * DoberMann[ZZZzzz] Ping timeout: 480 seconds 1186102435 Q * gerrit Ping timeout: 480 seconds 1186103054 J * brcc_ bruce@72.20.27.65 1186103073 M * brcc_ I need an opinio. Opteron or xeon ? 1186103086 M * daniel_hozac yes :) 1186103094 M * brcc_ Daniel! :) 1186104766 J * gerrit ~gerrit@c-67-169-199-103.hsd1.or.comcast.net 1186107033 J * FireEgl FireEgl@4.0.0.0.1.0.0.0.c.d.4.8.0.c.5.0.1.0.0.2.ip6.arpa 1186112386 Q * zLinux Ping timeout: 480 seconds 1186112442 J * zLinux ~zLinux@88.213.32.248 1186115002 J * Greek0 ~greek0@85.255.145.201 1186115179 Q * meandtheshell Ping timeout: 480 seconds 1186115336 Q * s0undt3ch Ping timeout: 480 seconds 1186116637 J * meandtheshell ~markus@85.127.108.206 1186121672 N * DoberMann_ DoberMann[PullA] 1186124172 M * eyck_ of course 1186124635 J * dna ~naucki@137-238-dsl.kielnet.net 1186125344 N * Bertl_zZ Bertl 1186125348 M * Bertl morning folks! 1186125368 M * coderanger Bertl: g'morning 1186125396 M * Bertl coderanger: don't worry, there is something here already :) 1186125410 M * coderanger Bertl: Eeeexcellent 1186125416 M * coderanger I have another question actually 1186125419 M * Bertl I just need to finalize it and give it some tests 1186125437 M * Bertl (probably done after the weekend) 1186125475 M * coderanger Can we (read: you) add a context flag/cap to prevent anything inside from ever migrating anywhere else 1186125480 M * coderanger Like a VXC_MIGRATE 1186125496 M * coderanger It would just get and'ed with CAP_SYS_ADMIN 1186125526 M * Bertl hmm, you should really not set CAP_SYS_ADMIN 1186125544 M * Bertl but there is already CAP_CONTEXT 1186125598 M * Bertl (which effectively prevents utilizing the vserver API) 1186125643 M * coderanger The goal is to be able to throw a possibly unknown userspace into a context in such a way that it almost doesn't need to know it has been virtualized. The benefit of this is any processes we run in the host context won't be visible to basically anything 1186125676 M * Bertl that is fine, but you should run it with the typical guest rpiviledges 1186125694 M * coderanger So lets say someone wants to run debian on the laptop, its init just gets spawed in context 2, and it gets full admin access (I'm sure their init script set the hostname and whatnot) 1186125705 M * coderanger we just don't want it going back to the host context 1186125757 M * Bertl hmm, you know that if you give too many caps, then this context can easily hack data from the host context? 1186125779 M * coderanger Data isn't the issue, it just needs to not be able to kill procs in the host context 1186125808 M * Bertl well, those processes are located somewhere (filesystem), no? 1186125819 M * coderanger The host context will just contain the security service and a few other very privileged things 1186125827 M * coderanger You mean the binary? 1186125835 M * Bertl yep 1186125862 M * coderanger Most of them, yes 1186125865 M * Bertl so disabling the security service is quite simple 1186125880 M * Bertl modify the binary, reboot, done 1186125894 M * neuralis heh, not so much 1186125900 M * neuralis the security service is started in initramfs 1186125905 M * neuralis which is the host context. 1186125912 M * coderanger The binary we really care about (antitheft daemon) is stored in the initrd, which is checked by the firmware before booting 1186125922 M * neuralis you can't modify the initramfs because you'll clobber a signature that the firmware checks. 1186125971 M * neuralis so really, as long as we can protect a process in the host context from getting killed, stopped, or otherwise influenced, we're fine. 1186126010 M * Bertl okay, but you have to make sure that no single binary in the host context is outside the signed (secure) initramfs 1186126038 M * Bertl and you also have to figure a way to restore a defect initramfs 1186126110 M * neuralis yeah, we understand that fuly 1186126112 M * neuralis *fully 1186126149 M * Bertl okay, removing CAP_CONTEXT from a non host context will disallow any context related manipulations 1186126162 M * neuralis perfect. 1186126169 M * coderanger Shiny 1186126247 M * neuralis Bertl: we've been thinking about flying you out for a week as we're now pretty deep into the integration work 1186126250 M * neuralis Bertl: how's your schedule? 1186126316 M * Bertl I have some dates till next tuesday, but after that I should have time 1186126335 M * neuralis next tuesday meaning the 7th or the 14th? 1186126345 M * Bertl the 7th 1186126363 M * Bertl so 8th++ should be fine 1186126367 M * neuralis okay, i'll send out some e-mails on our end and figure it out 1186126392 M * neuralis coderanger: when do you leave? 1186126406 M * neuralis the 24th or something? 1186126430 M * coderanger neuralis: I leave the 24th, so the 23rd-ish will be my last day (gotta fit packing and all that in somewhere) 1186126452 M * neuralis right 1186126471 M * neuralis coderanger: do you think it'd be helpful to bring herbert out around the 8th or a week or so from then? 1186126525 M * coderanger neuralis: Its up to you and your schedule, any time between now and FRS is good from my end really. 1186126560 M * coderanger I haven't really be working with vserver as much as Michael though, so you may want to check with him 1186126565 M * Bertl if it is required, I probably can stay for two weeks 1186126592 M * coderanger neuralis: More (good) hands are rarely a bad thing though ;-) 1186126595 M * neuralis ok, i'll talk to michael and we'll see how to go about it 1186126696 J * yarihm ~yarihm@whitehead2.nine.ch 1186126706 M * coderanger neuralis: Wanna get woodhouse out here too so we can just have a powwow :) 1186126754 M * Bertl btw, did david include the inode accounting as discussed? 1186126774 M * neuralis i don't know. i have a phone call with him later today (friday) and will ask. 1186126801 M * eyck_ I think the date on ML is off by a week, I'm receiving mails dated "Date: Fri, 20 Jul 2007 15:55:40 +0200 1186126806 M * yarihm i'm running into Checksum errors for rpms util-vserver is fetching when bootstrapping a centos4 vserver ... i guess the information contained in the util-vserver-package is not quite up to date, does anyone know what files I'd have to update in /usr/lib/util-vserver/distributions/centos4 order to get that working? 1186126810 M * neuralis bertl: had you guys agreed on a way to proceed when you last spoke, or what's the next step? 1186126817 M * yarihm daniel_hozac maybe? :) 1186126826 M * eyck_ ~2 weeks 1186126832 M * Bertl neuralis: basically we agreed on postponing it with some ideas :) 1186126854 M * Bertl neuralis: but I guess it should get back into focus RSN 1186126861 M * neuralis yeah, asap 1186126869 M * neuralis trial-3 freeze is in <2wks 1186126882 M * neuralis and we'd like that to be the feature freeze date 1186126920 M * Bertl how full is his schedule atm? 1186126937 A * coderanger arms the anti-avian-pig gun .... 1186126963 M * Bertl :) 1186126977 M * harry i have kipmi0 process gone a bit ... loose... 1186126983 M * harry any ideas what could cause that? 1186127000 M * Bertl ipmi? 1186127036 M * harry 1262 root 34 19 0 0 0 S 1 0.0 11:34.44 kipmi0 1186127042 M * harry top - 03:47:08 up 16:06, 1 user, load average: 0.85, 0.71, 0.63 1186127067 M * harry it's a completely idle server with 4 dual core hyperthreaded xeon machine with 16GB of ram... 1186127075 M * harry it shouldn't have such a "high" load when idling 1186127173 M * Wonka load has nothing to do with cpu activity 1186127195 M * Wonka it's about the number of processes ready to run 1186127221 M * Wonka ready to be scheduled 1186127260 M * harry i know 1186127265 M * harry there's nothing running 1186127291 M * coderanger your shell? 1186127305 M * Bertl nad 20+ kernel threads :) 1186127308 M * Bertl *and 1186127333 M * coderanger system load is only useful as a relative indicator most of the time 1186127367 M * harry coderanger: no kiddin'... 1186127380 M * harry the only thing i see running, is that kipmi0 thing 1186127383 M * harry which is a kernel thread 1186127392 A * harry runs iostat 1186127501 M * harry nothin 1186127511 M * harry only a shitload of nb1-16 devices 1186127524 M * harry hmm... nb0-15 ;) 1186127531 M * harry why, who, where? 1186127869 M * yarihm hi everyone anyway =) ... sorry 1186127882 M * Bertl hey yarihm! 1186128057 M * yarihm no precompiled vserver-kernel available yet for fedora 7, is that correct? 1186128069 M * Bertl I doubt it 1186128129 M * yarihm you doubt that there is one or you doubt that this is correct? 1186128141 M * yarihm (the latter I hope) 1186128166 M * Bertl yes, I doubt that daniel_hozac didn't build any vserver kernels for fc7 1186128193 M * coderanger Ours is technically an FC7 kernel :) 1186128200 M * yarihm Bertl: watch out, fedora-buys will tell you that it is _not_ fc7, as fedora 7 is not a core release anymore 1186128203 M * yarihm >( 1186128205 M * yarihm :) 1186128216 M * Bertl okay, f7 1186128243 M * yarihm well, in the repo of f7 there is none according to yum search and on dhozac's site there is no repo for f7 ... 1186128246 M * yarihm let me check again 1186128259 M * Bertl yarihm: double check with him 1186128284 M * yarihm I think he's asleep at the moment or busy otherwise ... idle since 6 hours 1186128312 M * Bertl yep, didn't say you should check right now :) 1186128318 M * yarihm I'll sure ask him when he's back :) 1186129896 J * Morsik ~mors@86.57.158.129 1186129899 P * Morsik 1186129908 M * Bertl short visit :) 1186129926 J * Piet ~piet@tor.noreply.org 1186130891 M * harry Bertl: time flies... he's trying to fly faster 1186132082 J * chand ~chand@212.99.51.254 1186133202 J * rgl ~rgl@84.90.10.107 1186133205 M * rgl hi 1186133401 A * rgl humm new releases :D 1186133413 M * Bertl hey rgl! 1186133427 M * rgl hello there :D 1186133435 M * rgl humm "use ULL instead of UL where appropriate" 1186133441 M * rgl what is UL? 1186133448 M * Bertl unsigned long 1186133464 M * rgl ULL is 64bit? 1186133480 M * Bertl unsigned long long :) 1186133487 M * rgl sure :D 1186133496 M * rgl but on a 64-bit machine, its 64-bit? 1186133520 M * rgl or even on a 32-bit machine 1186133529 M * Bertl that depends on the arch :) 1186133550 M * rgl is there anything but x86? heheh 1186133562 M * Bertl yes :) 1186133589 M * rgl I known :D but for my case, I never used anything but the x86 thingies. 1186133592 M * eyck_ hmm, i thought unsigned long long is at least 64bit on any arch... 1186133612 M * rgl oh, I used arm too. but for embedded stuff. 1186133638 M * Bertl eyck_: correct 1186133693 M * eyck_ good. 1186134709 M * rgl on my amd64: 1186134711 M * rgl sizeof int=4 1186134711 M * rgl sizeof long=8 1186134711 M * rgl sizeof long long=8 1186134724 M * rgl (bytes) 1186134727 M * Bertl congrats! 1186134743 M * rgl for what? hehe 1186134748 J * DavidS ~david@p57A497C5.dip0.t-ipconnect.de 1186134755 M * Bertl now check with linux32 prepended :) 1186134769 M * rgl what is linux32? :D 1186134800 M * Bertl part of the setarch package/utils 1186135167 M * rgl linux32 ./a.out 1186135167 M * rgl sysname=Linux 1186135167 M * rgl nodename=rgl-desktop 1186135167 M * rgl release=2.6.20-16-generic 1186135167 M * rgl version=#2 SMP Thu Jun 7 19:00:28 UTC 2007 1186135168 M * rgl machine=i686 1186135170 M * rgl sizeof int=4 1186135172 M * rgl sizeof long=8 1186135174 M * rgl sizeof long long=8 1186135184 M * rgl it should have changed Bertl ? 1186135220 J * bonsaikitten dreeevil@dev.gentooexperimental.org 1186135252 M * bonsaikitten hey - got one of those RTFM questions ;-) 1186135265 M * rgl the only thing that changed is the machine line (normally its x86_64) 1186135276 M * bonsaikitten do vservers support IP-Multicast? Or rather, is there anything that would cause problems? 1186135337 M * rgl I dunno. but did you try it? :D 1186135383 M * bonsaikitten not yet, but I thought I might ask before I spend a day or two fiddling with the impossible ;-) 1186135401 M * bonsaikitten I assume that it should just work, as most things in linux-vserver these days 1186135425 M * rgl humm, have you ever tried it in a non-vserver (in normal linux)? 1186135434 M * bonsaikitten yes, we have a setup here 1186135443 M * bonsaikitten don't know the details, but "it works" 1186135471 M * rgl can't you just copy it inside a vserver guest, and try it? 1186135499 M * bonsaikitten yeah, can do, will just take some time to get it all set up 1186135512 M * bonsaikitten don't worry, I'll just poke it until it works then ;-) 1186135563 M * rgl I never tried multicast, but from your reaction, it seems the setup is rather long? 1186135824 M * bonsaikitten the tools are a bit ... special :-) 1186135830 M * bonsaikitten and I haven;t set it up yet 1186136262 M * Bertl daniel_hozac: the existing ipv6 patches added the 'only add if new' functionality for v4 and v6, do you need/want that for anything? 1186136479 J * ensc|w ~ensc@www.sigma-chemnitz.de 1186137184 Q * lilalinux_ Remote host closed the connection 1186138244 Q * Piet Ping timeout: 480 seconds 1186138788 J * Piet ~piet@tor.noreply.org 1186138859 M * Borg- Bertl: can you check one thing for me please? 1186138865 M * Borg- nc -u -l -p 2500 1186138880 M * Borg- nc -u 127.0.0.1 2500 on same guest and type few times on both dirrections 1186138915 M * Borg- does it work fine for you? I get ICMP Unreach efter second packet from when im typing on nc -u 127.0.0.1 2500 1186138924 M * Bertl which kernel? 1186138974 M * Borg- Im using 2.6.17.13.. and I have problem in our production.. so I consider an upgrade 1186139057 M * Bertl looks fine here 1186139083 M * Bertl maybe you could do a test bash script which triggers this for you? 1186139137 M * Borg- hmm let me see. 1186139144 M * rgl is there an way to make an sparse file? I want to put a guest inside a file (null mount it?), but I don't want to allocate the file at once. 1186139172 M * Bertl rgl: dd? 1186139184 M * rgl or a dd from /dev/zero will make a sparse file? 1186139204 M * Bertl well, not if you dd the entire file of course :) 1186139205 M * eyck_ no 1186139221 M * rgl so how can I do it? 1186139257 M * Bertl dd if=/dev/zero of=sparse.file seek=1000 bs=1M count=1 1186139314 M * rgl ah, just seek to the EOF thx :) 1186139337 M * Bertl rgl: but note, that putting a filesystem on that will make the file quite real really soon :) 1186139361 M * rgl is there a way to see which parts of the file aren't allocated? 1186139383 M * Bertl yes 1186139390 M * rgl how? :) 1186139413 M * Bertl I was somewhat hoping you won't ask :) 1186139487 M * rgl no problem you don't remember, I'll try to look it up hehe 1186139504 M * Bertl it's not that I do not remember, but it is non-trivial 1186139548 Q * nebuchadnezzar Quit: ERC Version 5.2 (IRC client for Emacs) 1186139562 M * Bertl rgl: if you are just interested in the actual space consumption 1186139570 M * Bertl then du is a viable solution 1186139655 M * rgl ah ok. its fine using du. though, I was curious to see the patterns of allocation. hehe 1186139718 Q * bonsaikitten Quit: byebye 1186139739 M * Bertl rgl: well, if you find a trivial way to 'map' that, please let me know 1186139771 M * rgl Bertl, isn't there some kind of "defrag" for linux/ext3? 1186139780 M * Bertl what for? 1186139806 M * rgl to see the allocation patterns 1186139826 M * rgl I mean, how stuff is distributed on disk. 1186139866 M * rgl like, somehow see how the inodes are dispersed on the system. 1186139877 M * rgl err on the FS/disk. 1186139882 M * Bertl you can get that with dumpe2fs 1186139890 M * eyck_ there is a universal userspace defragmenter for unix filesystems.. 1186139946 A * rgl looks for dumpe2fs :D 1186139954 M * rgl eyck_, which is? 1186140033 J * nebuchadnezzar ~nebu@zion.asgardr.info 1186140034 M * eyck_ http://ck.kolivas.org/apps/defrag/ or http://vleu.net/shake/ 1186140044 M * rgl Bertl, dumpe2fs is nice :D known a graphical wrapper for it? :D 1186140066 M * Bertl rgl: never needed one :) but maybe google knows? 1186140067 M * eyck_ huh, graphical, huhu, wrapper. 1186140109 M * rgl ok thx :) 1186140212 M * rgl http://davtools.sourceforge.net/dav2.png :D 1186140245 M * Bertl fascinating :) 1186140491 Q * FireEgl Read error: No route to host 1186140582 M * rgl unfortunately there is not ubuntu package for it :| 1186140600 M * rgl oh, and it needs gtk1 :| 1186140731 M * rgl oh, it can use GTK2 sweet :D 1186140912 Q * DavidS Quit: Leaving. 1186141724 M * rgl you guys use vserver in the desktop? 1186141738 M * Borg- Bertl: are you there? I wrote very simple test 1186141765 M * rgl I'm planning in running it in my desktop, where I run the normal X in the host, but for testing and stuff, I'll make guests. 1186141785 M * eyck_ yes. 1186142036 M * Borg- anyone w/ newer than VS 2.6.17.13 is willing to help? 1186142046 M * Borg- I have simple script to test UDP loopback rewriting 1186142065 M * Borg- on 2.6.17.13 something is broken.. I receive only one packet 1186142172 M * Borg- http://borg.uu3.net/~borg/temp/udptest.sh 1186142180 M * Borg- run it 2 times.. on guest. 1186142185 M * Borg- 1) use guest IP 1186142190 M * Borg- 2) use 127.0.0.1 1186142196 M * Borg- use any not used UDP port 1186142205 M * Borg- tell me your results 1186142246 J * FireEgl FireEgl@Sebastian.Atlantica.IRCNut.Com 1186142290 M * yarihm daniel_hozac: around? 1186142361 M * daniel_hozac Bertl: i'm fine with removing that (the only add if new functionality). 1186142366 M * daniel_hozac yarihm: am now. 1186142376 M * yarihm daniel_hozac: cool :) 1186142378 M * yarihm just in time 1186142403 M * daniel_hozac Bertl: seems the old list started working again? 1186142454 M * Borg- so?? anyone? 1186142527 Q * [PUPPETS]Gonzo Quit: Serverwechsel 1186142553 M * yarihm daniel_hozac: i'm running into an error when trying to bootstrap a centos4-vserver: 1186142562 M * yarihm http://mirror.centos.org/centos/4/addons/i386/repodata/primary.xml.gz: [Errno -1] Metadata file does not match checksum 1186142572 M * yarihm it then continues to try with other mirrors until it fails eventually 1186142606 Q * gerrit Remote host closed the connection 1186142609 M * daniel_hozac Borg-: /dev/udp doesn't work here, so i can't test it :) 1186142622 M * yarihm where is that checksum defined so i can "tune" it a bit? i tried with several distributions/centos4-directories from centos5 to fedora 7 and debian sid ... all show the same error 1186142624 M * Bertl daniel_hozac: yeah, probably after martin sent me the subscriber list ... he now got the time to fix it 1186142665 M * daniel_hozac yarihm: in repomd.xml 1186142677 M * daniel_hozac yarihm: that's the first file it fetches from the repository. 1186142695 M * yarihm daniel_hozac: so in some sense, the repository is not consistent in itself? 1186142728 M * daniel_hozac yarihm: well... more likely you're just using different mirrors for the different files, one of which isn't updated. 1186142787 M * yarihm daniel_hozac: well "I'm just using ... " :) i just do vserver centos4 build -m yum... however, I can't find said file anywhere 1186142816 J * [PUPPETS]Gonzo gonzo@langweiligneutral.deswahnsinns.de 1186142843 M * daniel_hozac yarihm: yes, the default repo files use the mirrorlists. 1186142892 M * daniel_hozac you could try using your own repo definitions where you specify a specific mirror. 1186142944 M * Borg- daniel_hozac: huh? what bash do you have? :)P 1186142964 M * daniel_hozac whatever comes with etch... 1186142981 M * Borg- huh.. Debian.. funny distro 1186142995 M * Borg- maybe ash is used instead of bash? 1186143012 M * yarihm daniel_hozac: that sounds good. that would be /usr/lib/util-vserver/distributions/yum.repos.d/CentoOS-Base then? 1186143049 M * daniel_hozac yarihm: well.... copy that directory (yum.repos.d) to /etc/vservers/.distributions/centos first. 1186143068 M * Borg- still util-vserver is on top ? 1186143101 M * Borg- Bertl: please try my script and tell me results. 1186143111 M * yarihm daniel_hozac: ok, and then edit it? 1186143128 M * Borg- or anyone w/ bash that is capable of network redirections /dev/udp/ 1186143143 M * Bertl Borg-: yes sir! :) 1186143269 M * Bertl Borg-: hmm, well, /dev/udp/127.0.0.1/2500: No such file or directory? 1186143281 M * Borg- Bertl: running etch aswell? 1186143291 M * Bertl yep, for the test :) 1186143304 M * Borg- geez.. ppl at Debian should get a clue :) 1186143324 M * Bertl so, work around? new bash? 1186143324 M * Borg- wait.. I will rewrite it for netcat totaly.. 1186143482 J * ema ~ema@rtfm.galliera.it 1186143491 M * Borg- okey done.. please redownload and try again 1186143572 M * Bertl hmm, interesting ... 1186143585 M * Bertl (read: yes, that does not behave as expected) 1186143591 M * Borg- hm :/ 1186143619 M * Bertl but that is actually good :) 1186143641 M * Bertl because now we can check why ... 1186143658 M * Bertl (will do so shortly) 1186143659 M * daniel_hozac works fine here... 1186143676 M * Bertl 2.6.21.6-vs2.2.0.2 here 1186143684 M * daniel_hozac 2.6.20 on that box. 1186143719 M * Bertl how many ips? 1186143744 M * yarihm umm, that question must seem stupid, but what do RHEL/Centos people use to install packages? i run into trouble intstalling yum and as I'm not a rpm-distro-person I don't know anything else 1186143758 M * Bertl daniel_hozac: (for the guest) and with 127.0.0.1? and source remap enabled or disabled? 1186143761 M * daniel_hozac just the one. 1186143778 M * daniel_hozac disabled. 1186143791 M * yarihm in other words: i must create a vserver for a customer (centos4) and yum kinda fails to work, what else should I try to install? 1186143801 M * daniel_hozac yum is what you use... 1186143818 M * Borg- daniel_hozac: works both for 127.0.0.1 and guest IP? :) 1186143823 M * daniel_hozac Borg-: yep. 1186143829 M * yarihm daniel_hozac: ok. this is the error I get: 1186143830 M * yarihm bash-3.00# yum update 1186143830 M * yarihm rpmdb: Program version 4.2 doesn't match environment version 1186143830 M * yarihm error: db4 error(22) from dbenv->open: Invalid argument 1186143841 M * daniel_hozac yarihm: when you do what? 1186143853 M * daniel_hozac yarihm: i.e. at which step of the process are you now? 1186143878 M * yarihm and so on ... is there a requirement missing? that's what happened after vyum vservername -- install yum ; vserver vservername pkgmgmgt internalize ; vserver vservername exec yum update 1186143906 M * daniel_hozac it just means you didn't rebuild your database with the right db4 version. 1186143915 M * daniel_hozac rm -f /var/lib/rpm/__db*; rpm --rebuilddb 1186144006 M * yarihm ok, trying ... one sec 1186144062 M * yarihm Cannot find a valid baseurl for repo: update 1186144062 M * yarihm Error: Cannot find a valid baseurl for repo: update 1186144075 M * yarihm some repodefinition is missing as it seems ... 1186144079 M * daniel_hozac does the guest have internet connectivity? 1186144158 M * yarihm daniel_hozac: you are my hero 1186144229 M * Borg- daniel_hozac: do you have any special settings on your guest? (for that UDP 127.0.0.1) ? 1186144241 M * daniel_hozac none whatsoever, just built it. 1186144274 M * Borg- daniel_hozac: 2.6.20.15 ? 1186144298 M * Borg- w/ util-vserver 2.2.0.3 ? 1186144309 M * daniel_hozac i think it's 2.6.20.11... i don't recallt exactly. 1186144330 M * Borg- can you uname -sr ? :) 1186144343 M * daniel_hozac Linux 2.6.20-1.2949.fc6.vs2.2.0.2 1186144363 M * daniel_hozac (note: that's not vs2.2.0.2, that's vs2.2.0, release 2) 1186144389 M * Borg- hmm.. okey.. we need some more ppl to test it.. its strange 1186144515 M * yarihm daniel_hozac: thank you a lot (once again) 1186144546 M * daniel_hozac yarihm: so everything is working now? 1186144576 M * yarihm daniel_hozac: yes ... what I wanted to say though is that vserver ... build "crashed" in the middle of the process, but the result is still usable 1186144640 M * yarihm whatever, unless you are interested in the error message there, everything works. thanks 1186144661 M * yarihm (at least yum and ssh work AFAICT) 1186144681 M * daniel_hozac what was the error? 1186144748 M * yarihm let me reproduce it 1186144939 J * s0undt3ch ~s0undt3ch@80.69.34.154 1186144946 Q * s0undt3ch 1186145103 M * yarihm daniel_hozac: it goes like this: (the topmost line is just an error of debian's yum-version) 1186145105 M * yarihm Warning, could not load sqlite, falling back to pickle 1186145105 M * yarihm RTNETLINK answers: Numerical result out of range 1186145105 M * yarihm /usr/lib/util-vserver/distributions/centos4/initpost: line 105: 5301 Killed $_VSERVER "$vserver" exec bash -c ': >/tmp/startwait' >&/dev/null 1186145120 M * daniel_hozac ah, well, that's fine. 1186145139 M * daniel_hozac (not the RTNETLINK message, that means something's up with the IP address you assigned) 1186145327 J * FHTech ~Miranda@62.140.244.26 1186145334 M * Bertl welcome FHTech! 1186145401 M * FHTech Hello, who can to help that small advice on installation vserver on debian etch ? 1186145475 M * Bertl let's hear what the problem is ... 1186145641 M * FHTech Thanks, I establish on help http://www.howtoforge.com/linux_vserver_debian_etch. But after reboot it is visible 2.6.18-4-686. 1186145643 M * Bertl daniel_hozac: you have the ipv6 patches in your kernel? 1186145649 M * FHTech In a file/boot/grub/menu.lst are registered title Debian GNU/Linux, kernel 2.6.18-4-686 and title Debian GNU/Linux, kernel 2.6.18-4-vserver-686 1186145679 M * rgl how do you guys start configuring the kernel? with make defconfig? allnoconfig? 1186145691 M * Bertl rgl: vi .config :) 1186145701 M * Guy- bzcat >.config :) 1186145707 M * Guy- (you can type faster :) 1186145709 M * Borg- Bertl: any clues w/ that UDP 127.0.0.1 issue? :) 1186145716 M * rgl Bertl, but which start .config do you use? 1186145718 M * Bertl Borg-: work in progress 1186145724 M * Borg- ok :) great 1186145729 M * Borg- thx man 1186145739 M * rgl Bertl, an empty file? :D 1186145742 M * Guy- rgl: I normally just use the .config of a previous kernel I built 1186145748 M * Bertl rgl: the one from /proc/config.gz or 'make defconfig' 1186145760 M * daniel_hozac Bertl: not in this one. 1186145798 M * Bertl daniel_hozac: okay, I can reproduce it here with 2.6.20 and 2.6.22 at least ... 1186145798 M * rgl I was using the ubuntu version of .config, but its huge (it builds everything) *G* 1186145814 M * Bertl rgl: yeah, better avoid that 1186145815 M * rgl Guy-, but how did you bootstrap it? 1186145816 M * daniel_hozac Bertl: REMAP? 1186145840 M * Bertl FHTech: okay, so you want to select the vserver kernel at boot 1186145864 M * Bertl daniel_hozac: disabled in both of them 1186145865 J * lilalinux ~plasma@dslb-084-058-217-006.pools.arcor-ip.net 1186145880 M * Guy- rgl: you mean, how did I build my first kernel? 1186145896 Q * lilalinux Remote host closed the connection 1186145916 M * Guy- rgl: that was in '95, I believe, and I just did 'make config' then :) 1186145948 M * rgl Guy-, so, from there on you've been using your old config file? updating using make oldconfig? 1186145971 M * Guy- rgl: more or less 1186145980 M * Guy- rgl: of course, I do make menuconfig occasionally 1186145998 M * Guy- rgl: and certainly it must have happened that I started with a vanilla kernel source and make menuconfig 1186146022 M * rgl ok :) 1186146035 M * Guy- rgl: but by and large, I have some .config files lying around for different kinds of computers and I tend to use these as starting points 1186146087 M * FHTech Bertl: kernel 2.6.18-4-686 it is loaded by default. If I shall remove from a file /boot/grub/menu.lst title Debian GNU/Linux, kernel 2.6.18-4-686 and Debian GNU/Linux, kernel 2.6.18-4-686 (single-user mode), I shall leave Debian GNU/Linux, kernel 2.6.18-4-vserver-686 and Debian GNU/Linux, kernel 2.6.18-4-vserver-686 (single-user mode) by default will be is loaded kernel 2.6.18-4-vserver-686? Or as it is possible to make loading by default 2.6.18-4-vserver-6 1186146105 M * rgl I'll start doing that now for my desktop. for the server I always starting with make defconfig followed by make menuconfig. 1186146130 M * Bertl FHTech: you can either select it at the boot console, or 'reorder' the entries, or just specify a different default 1186146144 M * Guy- rgl: frankly, I don't see what difference it makes - you should arrive at the same config in the end, it's just a question of how much time it takes 1186146147 M * Bertl FHTech: I presume there is a line called 'default 0' 1186146159 Q * FloodServ synthon.oftc.net services.oftc.net 1186146159 M * Bertl FHTech: changing that to 'default 1' should do the trick 1186146228 M * rgl Guy-, yes, I'm trying to take less time configuring this. hehe 1186146229 M * Bertl daniel_hozac: basically confirmed it on all kernels so far, could you upload your kernel for me (just for a test)? 1186146246 M * Bertl daniel_hozac: assuming that it doesn't heavily depend on modules :) 1186146274 M * Guy- rgl: well, make oldconfig is bound to take the least time :) 1186146279 N * BobR_afk BobR 1186146542 N * BobR BobR_afk 1186146606 M * daniel_hozac Bertl: http://rpm.hozac.com/dhozac/fedora/6/vserver/i386/kernel-2.6.20-1.2949.fc6.vs2.2.0.2.i686.rpm 1186146643 M * daniel_hozac (yes, it depends heavily on modules. it's a distribution kernel :P) 1186146679 M * Bertl hmm, could you build one for me with a config I provide? 1186146690 M * Bertl (easily I mean :) 1186146717 M * daniel_hozac sure... want the same gcc and such? 1186146780 M * Bertl I don't think it is a compiler issue 1186146800 M * Bertl so whatever you ahve should be fine, just x86 and the config I provide 1186146885 M * daniel_hozac sure. 1186146888 M * Bertl daniel_hozac: http://vserver.13thfloor.at/Stuff/qemu-2.6.20.15-vs2.2.x.config 1186146924 M * harry # Linux kernel version: 2.6.20.14 1186146932 M * harry you sure it's 2.6.20.15 config? ;) 1186146953 M * Bertl yep :) 1186146993 M * daniel_hozac gonna take a while, seems i don't have any recent build trees around on that box. 1186146997 M * harry Bertl: what io scheduler do you normally use for web/db servers/ 1186146997 M * harry ? 1186147003 M * Bertl daniel_hozac: np 1186147009 M * harry cfq, deadline? 1186147016 M * Bertl cfq usually 1186147017 M * harry does it make a big difference? 1186147127 M * Bertl depends 1186147154 Q * FHTech Quit: FHTech 1186147162 M * harry i mostly use deadline 1186147180 M * harry since i only run servers, it doesn't matter all that much 1186147188 M * harry but maybe it gives an unfair result for vservers 1186147235 Q * rgl Quit: Zzzzz 1186147299 J * arcil ~arcil@p5B07660D.dip.t-dialin.net 1186147311 M * Bertl welcome arcil! 1186147324 J * rgl ~Rui@84.90.10.107 1186147352 M * rgl you guys use the "Build a relocatable kernel" feature of the kernel (2.6.22)? 1186147359 M * rgl (its enabled by default) 1186147383 M * Bertl of course, and we relocate our kernel on a regular basis :) 1186147417 M * eyck_ hmm, from one geographical area to another? 1186147422 M * harry it's allways fun to have a kernel here or there now and then 1186147426 M * rgl sometimes I have trouble figuring if you are joking Bertl ! 1186147427 M * eyck_ that's even better then checkpointing... 1186147431 M * harry especially if your kernel doesn't start at NULL 1186147451 M * Bertl eyck_: if you don't do that, then memory gets moldy ... 1186147475 M * harry if NULL is userspace, it would be very nice to mmap a page there and exploit your favourite nullptr derefs : 1186147478 M * harry ; 1186147480 M * harry ;) dammit! 1186147491 M * eyck_ we have improved our ventilation systems lately, we're no longer affraid of moldy memory 1186147752 Q * Aiken Remote host closed the connection 1186148060 Q * arcil Remote host closed the connection 1186148326 M * daniel_hozac Bertl: do you just want the kernel, or should i try to run it? 1186148351 M * Bertl well, for a start, the kernel would be fine :) 1186148370 M * Bertl if I encounter unusual issues with it, we consider other options :) 1186148427 M * daniel_hozac http://people.linux-vserver.org/~dhozac/bzImage 1186148541 M * Bertl okay, I get the same issue with this one 1186148683 M * Bertl what relevant differences do we have in the config? 1186148878 M * daniel_hozac what the. my kernel apparently has REMAP_SADDR enabled. 1186148911 M * Bertl okay, should be easy to test 1186148918 M * daniel_hozac i have no idea how that happened... 1186149228 M * bXi is it possible to have a server with 1 outside ip 1186149238 M * bXi which has a domains catchall pointed to it 1186149249 M * bXi to access the sepperate vservers trough the hostname? 1186149314 M * Bertl via http? yep, that is called 'proxy' 1186149372 M * bXi no on lower level 1186149378 M * harry bXi: impossible 1186149389 M * harry you mean: connect 1186149394 M * harry that resolves to ip 1186149395 M * Bertl bXi: depends on the protocol 1186149400 M * harry which connects to ip 1186149406 M * harry but then all hostname info is gone 1186149412 M * harry so you can't "filter" on that 1186149430 M * harry with webrequests, the name is still in your request, so you can filter on that 1186149435 M * harry for "normal" traffic, you can't 1186149459 M * bXi does ssh support this? 1186149470 M * harry you connect to an IP, the name-part is just to make it easy 1186149472 M * harry bXi: no 1186149532 M * bXi ipv6 it is then i guess 1186149555 M * harry yup :) 1186149609 M * bXi now to figure out how that dns stuff works exactly 1186149647 M * harry bXi: all computers are identified by there ip adress (all computers on internet) 1186149657 M * harry it's their "address" 1186149665 M * harry x.y.z.u 1186149668 M * bXi i know what dns does 1186149673 M * bXi but how to set it up 1186149674 M * harry mkay 1186149677 M * harry man bind ;) 1186149684 M * bXi i've got an /48 ipv6 range 1186149687 M * bXi a domain 1186149699 M * harry bloody much! :) 1186149702 M * bXi and an account at xname.org for dns managing 1186149726 M * bXi and a box with vservers 1186149882 M * harry Bertl: do you have any idea on how to monitor what kipmi0 is doing? 1186149892 M * harry it's taking up 1 min of cpu time every hour 1186149903 M * harry i know it 's not much etc... but it bothers me 1186149908 M * harry it shouldn't do anything 1186149925 M * harry on my (identical) other server, it took 7 minutes in 3 weekd 1186149927 M * harry weeks 1186149937 M * harry only difference: 2.2.0 -> 2.2.0.3 1186149940 M * harry for vserver 1186150259 J * freax ~linus@bl7-143-74.dsl.telepac.pt 1186150316 M * Bertl harry: (k)gdb? 1186150424 Q * Piet Ping timeout: 480 seconds 1186150443 J * FloodServ services@services.oftc.net 1186150668 Q * linus Ping timeout: 480 seconds 1186150750 M * Bertl daniel_hozac, Borg-: yep, enabling the saddr rewrite makes it work 1186150760 M * daniel_hozac okay. 1186150768 M * daniel_hozac that's... strange. 1186150986 M * rgl should I enable the "UTS Namespaces" on linux 2.6.22? 1186151007 M * Bertl probably a good idea :) 1186151011 M * daniel_hozac can you disable them on a Linux-VServer kernel? 1186151017 M * daniel_hozac because that would be a bug... 1186151023 J * arcil ~arcil@p5B07660D.dip.t-dialin.net 1186151036 M * Bertl wb freax! welcome arcil! 1186151045 Q * FireEgl Read error: Connection reset by peer 1186151055 M * rgl daniel_hozac, oh, I didn't yet apply the vserver patch to the kernel. maybe thats why I can change this. 1186151068 M * daniel_hozac yep. 1186151108 M * ema Guys, is there a way to see CONTEXT in vtop? 1186151115 M * daniel_hozac no. 1186151125 M * daniel_hozac vtop is just a wrapper for chcontext --xid 1 top 1186151152 M * ema mmmh, okay, and vps is not a wrapper for chcontex --xid 1 ps 1186151160 M * daniel_hozac nope. 1186151160 M * rgl OT: when I want to enable the hibernate/suspend thing on linux, I should create a swap big enought for all the mem? 1186151185 M * ema daniel_hozac: thanks 1186151186 M * daniel_hozac will be hard to fit it in there if you don't :) 1186151212 M * rgl oh, but that does not make sense, because when the swap is full, where will the suspend thing store its data? 1186151257 M * daniel_hozac suspend doesn't use swap. 1186151268 M * daniel_hozac hibernate will likely not store swap at all :) 1186151273 M * rgl what does it use? 1186151285 M * Bertl rgl: keep an unused swap space for swsusp and enable it shortly before suspend to disk 1186151336 M * rgl I had enable the suspend on gnome. the PC suspended, I clicked a key, it started to unsuspend, but, crashed :| 1186151404 M * daniel_hozac suspend won't cut the power to RAM.. 1186151448 M * rgl humm, the power will be on ram only? 1186151470 M * daniel_hozac well, and whatever wake-on devices you have enabled. 1186151540 M * harry is there a dietlibc for ubuntu 606 LTS ? 1186151564 M * daniel_hozac may be, but you probably don't want to use that. 1186151578 M * daniel_hozac dietlibc has been broken in most (all?) Ubuntu releases. 1186151588 M * harry ah 1186151615 M * harry i'm installing vserver on a ubuntu right now 1186151623 M * harry so better not use dietlibc then :) 1186151647 M * daniel_hozac well, you'd better use dietlibc, just not the Ubuntu one. 1186151681 M * rgl harry, I used the one from linux-vserver site, like daniel_hozac said told me before. 1186151702 M * harry Use dietlibc: no (you have been warned) 1186151707 M * harry so ignore that? 1186151713 M * harry aha 1186151719 M * rgl harry, just build it from http://svn.linux-vserver.org/svn/dietlibc ? 1186151892 M * harry i want packages! 1186151963 M * daniel_hozac use a sane distribution? 1186151964 M * daniel_hozac :) 1186151974 M * harry ubuntu lts not good? 1186151983 M * harry what's a good lts distro then? 1186151999 M * daniel_hozac Debian? :) 1186152025 M * sid3windr debian's dietlibc is broken too isn't it? :P 1186152033 M * daniel_hozac just for sparc64, no? 1186152046 J * FireEgl FireEgl@Sebastian.Tcldrop.US 1186152046 M * sid3windr no idea 1186152065 M * daniel_hozac i've not heard any of any other problems from Debian-users. 1186152072 M * daniel_hozac -any 1186152079 M * rgl harry, well, I'm using ubuntu too. and building the tools is easy, want my recipe? 1186152130 M * harry rgl: i just need the dietlibc :) 1186152139 M * harry but... i'm gonna reinstall teh entire server btw :) 1186152158 M * rgl harry, well, you'll have to build the tools with that dietlibc. 1186152178 M * harry i think i'll start with a kernel first 1186152182 M * harry then reinstall 1186152186 M * harry then i'll see what to do ;) 1186152200 M * rgl ok. if you need some help, just ask. 1186154120 J * DavidS ~david@p57A497C5.dip0.t-ipconnect.de 1186154393 Q * meandtheshell Ping timeout: 480 seconds 1186154725 J * FHTech ~Miranda@62.140.244.26 1186154885 M * FHTech hello, who can help with a question on OpenVCP, to be exact on IPFM. Parameter LOG IP/mask 1186154916 M * Bertl probably the maintainers of the project? 1186155133 J * fatgoose ~samuel@76-10-147-136.dsl.teksavvy.com 1186155217 M * FHTech Thanks, I have understood the mistake :) 1186155226 Q * FHTech Quit: FHTech 1186155478 J * meandtheshell ~markus@85.127.117.19 1186155576 J * stefani ~stefani@tsipoor.banerian.org 1186155579 Q * stefani 1186155677 Q * Baby Quit: KVIrc 3.2.0 'Realia' 1186155734 M * Bertl okay, dinnertime ... back later ... 1186155740 N * Bertl Bertl_oO 1186155757 J * lilalinux ~plasma@dslb-084-058-217-006.pools.arcor-ip.net 1186156508 J * Baby ~miry@195.37.62.208 1186157012 N * ensc Guest14 1186157022 J * ensc ~irc-ensc@p54B4D2C8.dip.t-dialin.net 1186157074 N * Bertl_oO Bertl 1186157078 M * Bertl back now ... 1186157129 Q * Guest14 Ping timeout: 480 seconds 1186157513 M * matti Hi B. 1186157523 M * Bertl Hi m. 1186157530 M * matti How's life? 1186157579 M * Bertl quite fine ... 1186157585 M * matti Good. 1186157590 M * matti At least one happy person! 1186157635 Q * lilalinux Remote host closed the connection 1186157758 Q * yarihm Quit: This computer has gone to sleep 1186157886 M * rgl can I use KVM with vserver? 1186157903 M * Bertl I hope so 1186158142 M * rgl http://virt.kernelnewbies.org/KVM/Performance 1186158157 M * rgl woah, the performance seems to drop considerably. 1186158231 M * Bertl between what and what? :) 1186158446 M * rgl running the benchmark program inside KVM guest vs KVM host. 1186158457 M * fatgoose ouch 1186158460 M * rgl then I guess, between vserver and kvm 1186158469 M * Bertl ah, well, expected :) 1186158495 M * rgl is xen this magnitude slower too? 1186158515 M * Bertl well, it is basically the same virtualization technology 1186158524 M * Bertl (give and take a few optimizations) 1186158541 M * rgl xen uses kvm at all? 1186158552 M * Bertl not yet 1186158555 M * rgl I mean, uses HW support? 1186158568 M * rgl so how can it be basically the same virt. tech? 1186158597 M * Bertl kvm != hardware support 1186158612 M * Bertl but, kvm _requires_ hardware support 1186158623 M * rgl ah! 1186158626 M * Bertl and xen (depending on the version) _uses_ hardware support 1186158639 M * rgl so what the HW support gives? 1186158649 M * Bertl another ring (-1) 1186158657 M * rgl just that? 1186158675 M * rgl I mean, is linux using all the four rings of x86? 1186158692 M * Bertl nope, only ring 0 and 3 1186158707 M * rgl so, kvm could be done without HW support? 1186158718 M * Bertl for linux guests, yes 1186158743 J * slack101 ~comp@cpe-71-72-115-74.insight.res.rr.com 1186158747 M * rgl humm, kvm runs other types of guests? 1186158757 M * rgl (non-linux that is) 1186158761 M * Bertl that's the basic idea behind it 1186158836 M * rgl but since it uses HW support, the guest does not need to be changed? 1186158846 M * slack101 Hello all 1186158846 J * bonbons ~bonbons@2001:960:7ab:0:20b:5dff:fec7:6b33 1186158864 M * rgl http://kvm.qumranet.com/ is dead here :( 1186159060 M * slack101 i just did a network benchmark between vserver and openVZ 1186159074 Q * chand Ping timeout: 480 seconds 1186159087 M * slack101 on deium load vserver wins 5 percent ......... on big load it went up to about 18 percent 1186159096 M * slack101 medium* 1186159134 M * rgl humm, openvz isn't virtualizing network adaptors? 1186159149 M * slack101 it virtulizes the network stack 1186159162 M * rgl ah, the whole shebang then. 1186159186 M * slack101 but on a small to medium load it wasnt really noticeable but yea 1186159230 M * rgl why do we need a virtualized stack? 1186159258 M * slack101 someone on the openVZ forum did it too ..........he tried ot bias them in his favor but vserver was about 10 percent better on his system at the load he did 1186159262 M * slack101 we? 1186159266 M * slack101 they do it on vserver 1186159269 M * slack101 not* 1186159297 M * rgl I mean, why did they virtualized the network stack 1186159299 M * rgl ? 1186159313 M * slack101 rgl: the upside is you can do iptabnles ipsec all that stuff ...........the downside is it is slower ....... 1186159314 M * rgl err what was the need to virtualize the network stack? 1186159334 M * slack101 ^^ 1186159336 M * Bertl rgl: it has some advantages 1186159361 M * Bertl rgl: and it simplifies things like snapshoting 1186159366 M * rgl slack101, but you can do it in the host, no? yeah, you need coordination between the guest and host on vserver. but I can live with that :) 1186159372 M * slack101 Bertl: speed is not one of them ;) 1186159380 M * Bertl slack101: that is correct 1186159391 M * slack101 rgl: i use a thing called virtuatables ....... kinda ghetto but works 1186159414 M * rgl slack101, what are those? 1186159432 M * rgl slack101, you use that on OVZ? 1186159438 M * slack101 run a daemon on the host ......... a user has a sh script he treats it like real iptables 1186159445 M * slack101 no no vserver 1186159457 M * slack101 i dunno some guy on a brazillian forum wipped it up 1186159481 M * slack101 although if you want it you will have to talk to daniel_hozac as he gave me the download 1186159498 M * rgl and the host will is periodically reading it? 1186159509 M * slack101 ummm 1186159511 M * rgl (and applying changes, I suppose) 1186159515 M * slack101 its a daemon 1186159528 M * slack101 accepts incoming connection -> do what it says 1186159554 M * rgl but how can it treat it like real iptables? 1186159566 M * slack101 it bounces it over 1186159572 M * slack101 relays w/e 1186159574 M * daniel_hozac it's just command-line compatible. 1186159598 M * rgl ah. its an iptables command that calls into the daemon running on the host using TCP? 1186159607 M * slack101 sure 1186159619 M * slack101 the user has a shell script 1186159685 M * rgl alright :D 1186159696 M * slack101 to communicate with the daemon but yea 1186159736 M * rgl woah, using kvm I can run unmodified windows and linux at the same time! 1186159748 M * slack101 -_- 1186159749 M * daniel_hozac yes, that's the idea.... 1186159765 M * rgl have you tried it? 1186159765 M * slack101 its not native speed rgl 1186159780 M * slack101 plus you need those amd extensions etc 1186159782 M * slack101 or intel 1186159795 M * rgl slack101, sure. but I do not need native speed for windows, just linux. 1186159808 M * slack101 plus it will cut down your system resources pretty good 1186159821 M * slack101 I am the virtulization king trust me i tried everything ;) 1186159832 M * rgl slack101, I got myself a C2D yesterday :D 1186159834 M * slack101 VirtualBox is gonna be the best you can get performance wise with windows 1186159843 M * slack101 better then KVM 1186159865 M * slack101 actually those super professional versions of vmware are the best 1186159885 M * rgl but using kvm with vserver on the host, and sometimes windows, there is no performance degradation, or is it? 1186159890 M * slack101 as i have heard anyways but yea .............stick to userland stuff 1186159900 M * rgl (when I'm not using the windows vm that is hehe) 1186159911 M * daniel_hozac rgl: shouldn't be. 1186159918 M * slack101 i disagree 1186159930 M * slack101 its virtulization 1186159939 M * slack101 vserver is pretty much native 1186159956 M * slack101 i ran 6 vservers and my hardware was doen for 1186159957 M * daniel_hozac vserver is not even in the same category 1186159957 M * rgl slack101, but most of the time I'm not running windows. 1186159976 M * rgl slack101, so, kvm is idle, no? 1186159980 M * slack101 i know daniel_hozac but thinking hes gonna get native perforamnce is a dream 1186159996 M * slack101 rgl: are you gonna provide windows vps ? 1186159998 M * daniel_hozac you know how small the KVM patch is, right? 1186160001 M * slack101 that was my reason 1186160010 M * rgl slack101, no. its for my personal desktop. 1186160017 M * slack101 well then that might be ok 1186160028 M * slack101 its fin if you got enough ram 1186160057 M * slack101 virtualbox is faster though execution wise ........it has a kernel module as well 1186160070 M * rgl slack101, 2G will do the trick? 1186160077 M * slack101 2G? 1186160087 M * rgl daniel_hozac, 2 GiB of ram. 1186160089 M * rgl errr 1186160094 M * rgl that was for slack101 . 1186160098 M * slack101 sure 1186160105 M * slack101 well i dont know what your going ot be running 1186160119 Q * ema Quit: leaving 1186160147 M * rgl slack101, nothing special. just printing stuff, mostly *G* 1186160162 M * slack101 then you can use qemu ..........i wouldnt worry about it 1186160170 M * rgl slack101, "linux" does not like my printer. 1186160174 M * slack101 i have a 800 mhz comp right here next to this one with a simple windows xp on it 1186160185 M * slack101 whats your printer ? 1186160198 M * rgl its a hp laserjet 1000 1186160210 M * slack101 pretty sure all that hp stuff is highly supported 1186160220 M * slack101 atleast my hp officejet was 1186160231 M * rgl last time I checked, it wasnt :( 1186160246 M * rgl it just prints at 600dpi... which is a waste of toner. 1186160280 M * slack101 http://www.linuxprinting.org/show_printer.cgi?recnum=HP-LaserJet_1000 1186160351 M * slack101 looks like that works fine 1186160433 M * rgl but it kinda sux. I tried hehe 1186160476 M * slack101 ah well 1186160492 M * slack101 how about buy a newer printer ? :D 1186160507 M * rgl why? the printer is fine hehe 1186160526 M * rgl the software drivers aren't thou... hehe 1186160615 M * rgl daniel_hozac, the kvm patch is big? 1186160620 M * slack101 not to start a flamewar but i used to work for a place that sold printer only ......... the laserjet 1000 was by far the worse one , i think it was released in 01 and was hp's first time to try n make a cheap laser printer ........it had so many returns 1186160644 M * Bertl Hollow: can we get a Gentoo:Linux-VServer Howto? 1186160644 M * slack101 the newer ones made by hp are good though 1186160669 M * rgl slack101, for my use, it fine. *G* 1186160676 M * slack101 :P 1186160682 M * rgl slack101, and yeah, it was cheap hehe 1186160691 M * Hollow Bertl: would a link to http://www.gentoo.org/proj/en/vps/vserver-howto.xml be enough? :) 1186160694 M * daniel_hozac rgl: no. 1186160701 M * daniel_hozac Hollow: can we do a redirect? ;) 1186160703 M * rgl slack101, I have it like, 4 or 5 years *G* 1186160722 M * Hollow daniel_hozac: if mediawiki supports external redirects.. 1186160737 M * Bertl Hollow: yep, looks good 1186160738 M * daniel_hozac that was essentially the question :) 1186160760 Q * DavidS Quit: Leaving. 1186160771 M * daniel_hozac hmm, isn't http://linux-vserver.org/Installation_on_Gentoo good enough? 1186160774 A * slack101 yawns 1186160801 M * rgl slack101, how does virtualbox compare with kvm? 1186160811 M * slack101 theres *many* benchmarks 1186160816 M * Bertl rgl: similar to QEMU 1186160821 M * slack101 vbox always wins by a slight margin 1186160836 A * rgl never used qemu 1186160836 M * slack101 Bertl: huh ????/ 1186160842 M * slack101 qemu is very slow 1186160848 M * slack101 i think he meant kqemu 1186160854 M * daniel_hozac uh. 1186160857 M * daniel_hozac qemu is just userspace. 1186160869 M * rgl or for the mater, I never used anything but vmware workstation on windows *G* 1186160871 M * daniel_hozac kvm and kqemu can both be used as acclereators for it... 1186160877 M * daniel_hozac +english 1186160887 M * slack101 qemu can be used without either 1186160894 M * daniel_hozac yes. ergo, accelerator. 1186160912 M * slack101 when someone says qemu that usually means qemu wiht no acceleration 1186160922 M * slack101 but anyways 1186160933 M * slack101 rgl: use w/e you will never know a difference 1186160935 M * rgl ""it is now the only professional-quality virtualization solution that is also Open Source Software.""" woah! 1186160952 M * slack101 rgl: who said that ? 1186160963 M * rgl so xen is amateur? and kvm? lol 1186160968 M * rgl http://virtualbox.org/wiki/VirtualBox 1186160971 M * slack101 oo 1186160975 M * slack101 vbox is a paid project 1186160982 M * slack101 they get much mcuh more done 1186160987 M * rgl so is xen? and kvm? 1186160987 M * daniel_hozac Xen isn't? 1186160995 M * slack101 i never looked into xen daniel_hozac 1186161033 M * daniel_hozac that's kinda strange for the virtualization king, no? 1186161041 M * slack101 no becasue 1186161049 M * slack101 at that time i ddint have the hardware for it daniel_hozac 1186161063 M * daniel_hozac huh? 1186161078 M * slack101 i did not have a processor with VT capabilities at that time 1186161096 M * slack101 and now I just dont care 1186161100 M * daniel_hozac KVM is the one requiring VT, not Xen. 1186161104 M * slack101 I use Vserver / Virtuzzo(windows) 1186161113 M * slack101 xen requires as well 1186161117 M * slack101 well it did 2 months ago 1186161119 M * daniel_hozac no, it doesn't. 1186161121 M * daniel_hozac it never has. 1186161132 M * slack101 *ONE* moment 1186161139 M * daniel_hozac i ran it on my 650 MHz P3 several months ago. 1186161148 M * daniel_hozac well, i tried. didn't really work out. 1186161149 M * rgl I think the xen guys said using the HW support was slower. 1186161191 M * slack101 i am 99.99 percent sure ot run windows with xen you need hardware support but 1 sec 1186161209 M * daniel_hozac to run unmodified Windows, yes. 1186161223 M * slack101 ? 1186161250 M * rgl you need HW support to run unmodified Windows 1186161270 M * slack101 what do we modify windows now ?????/ 1186161276 M * rgl a xen guest normally has to be rebuild to the xen arquitecture. 1186161297 M * rgl which you can't do with windows, I guess hehe 1186161304 M * slack101 anyways my point still stands at the time when i looked at it ...........no VT processor == no windows with xen 1186161336 M * rgl you are right slack101. it still stands now 1186161344 M * daniel_hozac so, question. if you didn't have VT, how did you look at KVM? 1186161354 M * slack101 daniel_hozac: please read up 1186161371 M * slack101 i bought one recently wit my new cpu daniel_hozac 1186161402 M * slack101 and to just add i got a server in a datacenter with one 1186161463 M * slack101 anyways ..... sorry i am in a bad mood :D 1186161470 M * daniel_hozac i don't care. 1186161473 M * eyck_ slack101: why we, microsoft and xen guys modified windows to run on xen 1186161477 M * slack101 i spent 10 hours debugging openvcp its a mess 1186161478 Q * FireEgl Read error: Connection reset by peer 1186161530 M * slack101 daniel_hozac: i was light weight aplogizing for my attitude, but w/e 1186161538 M * slack101 im sure xen is great i just really never tried it 1186161684 M * Bertl slack101: regarding qemu without acceleration: then you have to talk about VMware also without the VMware kernel module :) 1186161711 M * Bertl (which will make it terribly slow too :) 1186161725 M * slack101 agreed 1186161732 M * Bertl Hollow: when can we expect the magic link/redirect? 1186161798 Q * Guy- Ping timeout: 480 seconds 1186161949 N * DoberMann[PullA] DoberMann 1186161960 Q * Hollow Ping timeout: 480 seconds 1186162265 J * Hollow ~hollow@proteus.croup.de 1186162288 M * Hollow Bertl: will look at it in a minute.. just need to finish the openvpn config .. 1186162307 M * Bertl np 1186162480 J * FireEgl FireEgl@2001:5c0:84dc:1:4:: 1186162907 J * Piet ~piet@tor.noreply.org 1186163419 N * DoberMann DoberMann[PullA] 1186164119 M * rgl jeebus, I've compiled linux 2.6.22, rebooted, and the system crashes on boot, and turns off the screen :-( 1186164135 M * rgl any ideia how I'm going to troubeshoot this? 1186164164 M * daniel_hozac serial console. 1186164168 M * daniel_hozac the answer to everything :) 1186164228 M * rgl I can see that some kernel messages passing by on the console, but the damn thing turns off the screen *G* 1186164268 M * Wonka serial consoles rock the house 1186164270 M * rgl humm. going to try that daniel_hozac :D 1186164288 A * rgl goes to try figuring this serial thingy hehe 1186165006 M * matti daniel_hozac, Wonka: netconsole ain't that bad too tho 1186165007 M * matti ;] 1186165025 M * Bertl but sending magic sysrq is a little trickier :) 1186165027 M * daniel_hozac kinda hard to debug networking problems with it though ;) 1186165044 M * daniel_hozac Bertl wins :) 1186165066 M * Bertl daniel_hozac: note: it does work (if you know how :) 1186165114 M * slack101 i use serial and net consoles 1186165121 M * daniel_hozac congrats. 1186165135 M * daniel_hozac Bertl: any pointers? 1186165135 M * slack101 net for normal use and serial for when things really get bad :D 1186165308 M * daniel_hozac Bertl: i'm not seeing anything obvious just looking at the code, and the documentation doesn't mention it... 1186165327 M * rgl how do I open up a serial port terminal in linux? I want to first test the lisk between the PCs :D 1186165347 M * Bertl daniel_hozac: probably a feature which was removed since I gave up on supporting/updating it :) 1186165404 M * daniel_hozac rgl: minicom 1186165448 M * matti Bertl: Yep it can be tricky ;p 1186165450 M * matti Hehe 1186165615 Q * arcil Quit: Leaving 1186165668 N * DoberMann[PullA] DoberMann 1186165773 M * rgl daniel_hozac, woah. so complicated that program! 1186165814 M * slack101 might sound stupid but is their anyway to halt a vserver ? 1186165815 M * Bertl rgl: that was for modems 1186165837 M * slack101 like deactivate 1186165839 M * Bertl slack101: vserver name stop ? 1186165853 M * slack101 well the user could start it back up hmmm 1186165868 M * daniel_hozac your users have root on the host? 1186165883 M * slack101 i was wondering if there was a way to disable it until i would re enable 1186165885 M * slack101 daniel_hozac: you know it 1186165898 M * slack101 no of course not ........ happens through a panel 1186165910 M * daniel_hozac so that's a problem for your panel to solve, no? 1186165913 M * matti rgl: See if you have ckermit or microcom. 1186165920 M * matti rgl: They're very simple. 1186165923 M * slack101 daniel_hozac: yesi guess i couls just deactivate their panel 1186165929 M * matti rgl: You can set agetty on ttyS0 as well. 1186166120 M * rgl microcom is not on ubuntu archive :( 1186166137 M * rgl ckermit also seems complicated :| 1186166156 M * slack101 not really 1186166169 M * rgl gee, windows has hyperterminal.. which you only need to set the com number, and the rate, and thats it :| 1186166177 J * Guy- VpFTQTAnHs@chardonnay.math.bme.hu 1186166201 M * matti rgl: Give me a second. 1186166224 M * daniel_hozac rgl: minicom is the same way... 1186166226 M * matti rgl: http://microcom.port5.com/ - that's the web page. 1186166407 M * matti rgl: I can give you this : 1186166408 M * matti file ./microcom 1186166408 M * matti ./microcom: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 2.6.0, statically linked, for GNU/Linux 2.6.0, stripped 1186166470 M * rgl matti, I've built it. 1186166481 M * rgl matti, now how can I use it? *G* 1186166486 M * matti I just compiled it for ya ;p 1186166493 M * matti http://deadbeef.info/tmp/linux/microcom.gz 1186166508 M * matti rgl: http://microcom.port5.com/ - read the web page ;] 1186166520 M * rgl matti, I do not have the ~! 1186166525 M * matti rgl: microcom is so simple. 1186166532 M * slack101 lol 1186166535 M * matti You don't have what? 1186166536 M * rgl matti, that a composite key on my keyboard. 1186166556 M * rgl matti, the tilde key. the $HOME shortcot character. 1186166560 M * matti I know. 1186166574 M * matti Everybody have ~ 1186166579 M * matti Try to press ESC then 1186166609 M * matti microcom -D/dev/ttyS 1186166611 M * rgl matti, my keyboard is portuguse. 1186166616 M * rgl err portuguese. 1186166626 M * matti And if you have 9600,8n1 try to press enter few times. 1186166633 M * rgl we use ~ has an accent. 1186166633 M * sid3windr rgl: screen 1186166638 M * matti If there's something you'll get it. 1186166640 M * sid3windr screen /dev/ttyS0 9600 1186166649 M * rgl sid3windr, oh! sweet! 1186166689 M * rgl screen works. so sweet. thx :) 1186166691 M * sid3windr :) 1186166698 M * matti sid3windr: Nice trick :) 1186166706 M * sid3windr screen rawks 1186166722 M * matti rgl: Anyway, you have microcom, screen... :) 1186166808 M * fatgoose what does screen /dev/ttyS0 ... ? 1186166819 M * matti fatgoose: Opening serial port. 1186166828 M * fatgoose oh 1186166837 M * eyck_ cu is nicer 1186166850 M * matti eyck_: ? 1186166919 M * Bertl daniel_hozac: hmm, is it intentional that chcontext does not detect that it is running on a non-Linux-VServer kernel? 1186166997 M * daniel_hozac Bertl: well, i guess it could do that... 1186167022 M * Bertl I just wondered about this output: 1186167023 M * Bertl chcontext: tools were built without legacy API support; can not continue 1186167035 M * Bertl (which might be a little misleading :) 1186167061 M * daniel_hozac the question is just, where would you add such a check? 1186167081 M * daniel_hozac chcontext isn't used for e.g. vserver ... 1186167094 M * Bertl well, a good hint would be that the syscall for the version number returns ENOSYS? 1186167111 M * daniel_hozac indeed. 1186167121 M * Bertl so I would expect _any_ syscall to detect that :) 1186167141 M * daniel_hozac and output "non-Linux-VServer kernel"? that's not too nice :) 1186167156 M * Bertl maybe Linux-VServer syscall not implemented? 1186167166 M * matti Not detected? 1186167172 M * Bertl or if you want to give a usable hint: 1186167178 J * er ~yakker@aegis.CS.Princeton.EDU 1186167184 M * Bertl Kernel API does not support Linux-VServer syscall :) 1186167191 M * Bertl *ABI 1186167199 M * Bertl wb er! 1186167204 M * daniel_hozac still, these are all library functions. 1186167206 M * sid3windr babi! 1186167212 M * waldi hmm? 1186167214 M * daniel_hozac i'm not sure i want them outputting anything. 1186167231 M * Bertl daniel_hozac: okay, I'm fine if it is handled in the tools 1186167241 M * er Bertl, thanks! 1186167261 M * Bertl daniel_hozac: just the legacy API seems misleading to me 1186167279 M * daniel_hozac Bertl: definitely. 1186167299 M * matti rgl: Whoooaooo... I am trying to find in Google, how to produce ~ when using portuguese keyboard layout... no luck so far ;] 1186167343 M * eyck_ matti: cu -l /dev/ttyS0 1186167352 M * rgl matti, the normal way is to press ~ and SPACE. but that does not work with microcom. 1186167377 M * matti rgl: Oh. 1186167394 M * Bertl rgl: what about AltGr-~ ? 1186167394 M * matti I thought that you don't cannot produce ~ at all. 1186167407 M * rgl normally I never use the ~ key on the command line *G* 1186167417 M * Bertl or whatever the right Alt is called there 1186167429 M * matti eyck_: bash: cu: command not found 1186167440 M * sid3windr just type the ~ followed by spacebar? 1186167446 M * sid3windr instead of an n or whatever :) 1186167447 M * matti Should work. 1186167458 M * sid3windr ~ is deadkey on my belgian keyboard too 1186167526 M * rgl sid3windr, read above :) 1186167553 M * sid3windr ah 1186167554 M * sid3windr okay. 1186167556 M * sid3windr =) 1186167562 M * sid3windr seem to have missed that line 1186167570 M * matti sid3windr: How dare ya! 1186167570 M * matti ;D 1186167579 M * matti sid3windr: No dinner today. 1186167584 M * sid3windr already had dinner 1186167587 M * matti Damn. 1186167590 M * matti ;] 1186168035 M * rgl you guys use GCC 4.1.1 to compile the kernel? enable the Force gcc to inline functions marked 'inline' ? 1186168065 M * rgl (actually, its gcc 4.1.2) 1186168090 M * eyck_ matti: apt-get install cu ;) 1186168511 N * DoberMann DoberMann[PullA] 1186168795 M * rgl omg. linux does not output anything to the serial console :( 1186168853 M * sid3windr ? :) 1186168914 M * daniel_hozac did you boot with console=ttyS0,9600n1? 1186168959 M * rgl daniel_hozac, with console=ttyS0,38400n8 1186168967 M * rgl n1? 1186169078 M * daniel_hozac uh, right. 1186169080 M * Bertl n1 would be funny :) 1186169091 M * daniel_hozac set your terminal to the same settings? 1186169156 M * rgl daniel_hozac, yes. I tried with screen before booting, and I even tried to append the exact same line to the working kernel, and it works. 1186169245 M * rgl I must be doing something really wrong for this not even showing the console on ttyS0 :( 1186169369 M * matti eyck_: ! 1186169370 M * matti ;p 1186169529 M * slack101 rgl: still cant get microcom to work ? 1186169572 M * rgl slack101, I'm using screen. 1186169585 M * rgl slack101, I didn't made it work. 1186169593 M * slack101 make* 1186169605 M * rgl slack101, though, screen is way simplier. hehe 1186169607 M * slack101 you in austria too ? 1186169616 M * rgl Portugal 1186169633 M * slack101 ah 1186169713 J * meebey meebey@booster.qnetp.net 1186169903 Q * meandtheshell Quit: Leaving. 1186170160 M * Bertl wb meebey! 1186170240 M * meebey hiya Bertl :) 1186170279 M * meebey Bertl: klogd became worse btw, now its hanging being defunct, so I had to modify the init.d script to not run it 1186170295 M * meebey Bertl: worse being newer versions 1186170324 M * meebey root 22697 0.0 0.0 1692 456 pts/0 S+ 21:41 0:00 \_ klogd -x 1186170324 M * meebey root 22698 0.0 0.0 0 0 ? Zs 21:41 0:00 \_ [klogd] 1186170330 M * meebey Bertl: ugly eh? 1186170372 M * Bertl what kernel? 1186170634 Q * Piet Ping timeout: 480 seconds 1186170859 M * rgl oh man. I'm lost. the last message I can see from the kernel is, "booting kernel" then bang. the screen turns off, but the box keeps turn on, but it seems freezed :( 1186170888 M * rgl and nothing goes to the serial console at all :| 1186170927 M * Bertl maybe your serial port is not properly enabled? 1186170942 M * rgl how do you properly enable it? 1186170948 M * Bertl maybe you are listening to the wrong port or have the baud rate wrong? 1186170969 M * Bertl rgl: you did test the serial communication first, right? 1186170980 M * Bertl rgl: what ports/rates did you use? 1186170986 M * meebey Bertl: 2.6.18 1186170988 M * rgl no. I tried it with the working kernel, and that one sends the messages correctly through the serial port. 1186171009 M * rgl I did try the connection first. and it works. 1186171016 M * Bertl rgl: yes, that's what I meant, what ports/rates did you use? 1186171017 M * rgl but not with my compiled kernel :| 1186171027 M * daniel_hozac rgl: you did enable serial console, right? 1186171033 M * Bertl rgl: also, did you compile in serial and serial console? 1186171043 M * rgl console=ttyS0,38400n8 1186171062 M * Bertl and you are using 38400n8 on the other end too? 1186171065 M * daniel_hozac i was referring to the kernel config, much like Bertl ;) 1186171066 M * rgl oh... I did compile th serial port in, but I don't remeber seeing a "serial console " option :| 1186171081 M * Bertl rgl: then you definitely want to look for that :) 1186171143 M * sid3windr heheh 1186171264 M * rgl oh, I have this "[*] Console on 8250/16550 and compatible serial port" 1186171293 M * rgl that is the "Serial Console", correct? 1186171297 M * Bertl yep 1186171341 M * daniel_hozac Bertl: could i get the network.h from 2.3 too? 1186171350 M * rgl gag. its not working for some reason then. oh, I have placed "(1) Maximum number of 8250/16550 serial ports" .. I'll revert that (1) to (4) like it was before :| 1186171382 J * ema ~ema@rtfm.galliera.it 1186171398 M * Bertl daniel_hozac: sure, let me upload the latest vcmd with the (preliminary) 2.3 headers 1186171413 M * daniel_hozac okay, cool. 1186171538 M * Bertl http://vserver.13thfloor.at/Experimental/TOOLS/vcmd-0.11.tar.bz2 1186171569 M * Bertl daniel_hozac: note: I didn't update the shiny yet, but we should remove the untested part from arm 1186171581 M * rgl does not work... after the "Kernel alive" message, the screen turns off :| 1186171598 M * rgl (and nothing goes to the serial port) 1186171604 M * Bertl screen means: the screen or the serial console? 1186171631 M * rgl "the screen" "the monitor" 1186171638 M * Bertl rgl: maybe you are passing 'quiet' in the kernel boot line? 1186171646 M * rgl nope. no quiet. 1186171652 M * rgl or splash 1186171667 M * rgl kernel /vmlinuz-2.6.22.1-rgl-desktop root=UUID=81e7dc2d-20ec-44fa-9038-5ebaeb52a158 ro console=ttyS0,38400n8 1186171673 M * rgl thats the full line. 1186171694 M * Bertl and you checked twice that your other end is set to 38400n8 :) 1186171714 M * rgl Bertl, yes. it works with that exact line when I boot the ubuntu kernel. 1186171746 M * Bertl okay, but your kernel boots successfully? 1186171751 M * rgl no 1186171760 M * Bertl ah, it fails to boot at all? 1186171764 M * rgl my kernel seems to hang for some reason 1186171776 M * Bertl probably because you used the ubuntu gcc 1186171779 M * rgl yes. it seems to boot at all. 1186171822 M * rgl humm. yes, I use it. 1186171844 M * rgl the ubuntu kernel itself isn't compiled with the same gcc? 1186171845 J * fatgoose_ ~samuel@76-10-147-136.dsl.teksavvy.com 1186171854 M * Bertl rgl: no idea :) 1186171878 M * rgl but why you say "ubuntu gcc", is it really that full of patches? 1186171893 M * rgl that make it produce borked code? :| 1186171913 M * Bertl well, we had some troubles with it 1186171944 M * neuralis it's not full of patches, it's just that the ssp patches are slightly broken in not really letting you turn the functionality off 1186171959 M * neuralis not sure if it got fixed recently, it was still a problem as recently as feisty iirc 1186171972 M * neuralis only the 4.0 series; using 3.3 will work fine 1186172059 M * rgl I'm using gcc-4.1.2 1186172160 M * bored2sleep anybody tried to stick vserver into linux-2.6.23-rc1? 1186172178 M * Bertl bored2sleep: not that I know of 1186172178 M * daniel_hozac AStorm. 1186172203 M * daniel_hozac though that was probably some git version before -rc1, IIRC. 1186172212 M * bored2sleep k 1186172285 Q * fatgoose Ping timeout: 480 seconds 1186172419 M * neuralis rgl: yeah, i meant anything >=4 1186172600 M * fatgoose_ hey, any idea/tools that I could use to log all process spawned ? 1186172614 M * fatgoose_ I think that grsecurity got it 1186172625 M * fatgoose_ but not all my kernels have grsec patch 1186172663 M * daniel_hozac systemtap. 1186172676 J * yarihm ~yarihm@84-75-109-39.dclient.hispeed.ch 1186172784 M * fatgoose_ woah, thanks man 1186172955 Q * ema Quit: leaving 1186172971 M * rgl woah. it booted! I've disabled a bunch of stuff... 1186172992 M * rgl now I'll have to enable one by one, till it crashes again :| 1186173191 M * slack101 lol 1186173199 M * slack101 rgl: explain to me ....... what are you doing 1186173199 M * slack101 ? 1186173210 M * slack101 compiling a kernel ? 1186173228 M * rgl yes. 2.6.22.1 to try out the vserver patch. 1186173242 M * rgl but I'm first trying to build the vanilla kernel with the patch. 1186173249 M * rgl err without 1186173256 M * slack101 are you doing this for fun or for production ? 1186173276 M * rgl its for my desktop. for now its for fun. 1186173288 M * slack101 why a lionux vserver on your desktop ? 1186173313 M * rgl because I want to try stuff without borking my desktop system. 1186173332 M * slack101 wtf 1186173340 M * rgl like, try my software in different linux dists. 1186173343 M * slack101 heh i really dont think linux vserver is ideal for that 1186173354 M * slack101 i would use vbox for that 1186173397 M * Bertl rgl: note: slack101 is a _real_ vbox fan, but doesn't like qemu :) 1186173405 M * slack101 ok or qemu 1186173414 M * slack101 i am just used to saying vbox 1186173436 M * slack101 Bertl: im not even really a fan i dont use it nor prolly ever will ;) 1186173446 M * slack101 Virtuzzo for windows ......... Vserver for linux 1186173448 M * slack101 need no more 1186173449 M * rgl slack101, no. this way, I'm also testing vserver. 1186173451 M * daniel_hozac you sure are promoting it actively... 1186173470 M * daniel_hozac personally, i think vserver is perfect for testing software in different distributions. 1186173476 M * slack101 daniel_hozac: well it is a tad better then qemu BUT im still not even trying to promote it 1186173503 M * slack101 daniel_hozac: you would say anything to disagree 1186173525 M * daniel_hozac that's just because i think you're wrong 99% of the time :) 1186173539 M * Bertl make that 99.99 :) 1186173545 M * slack101 yea sure use os based isolation to system all your software 1186173571 M * slack101 nope its just matter of personal opinion 1186173587 M * slack101 daniel_hozac: you like sit back and wait to hound me on something 1186173595 M * slack101 but anyways doesnt matter 1186173602 M * daniel_hozac yes, i stalk you. 1186173615 M * slack101 if one is using os level baeed virtulization too test software they need help 1186173633 M * Bertl why? that's the best way to do it ... 1186173635 M * slack101 you would need a full virtulization 1186173645 M * slack101 Bertl: ever distro has a different kernel 1186173646 M * slack101 usually 1186173652 M * slack101 modified to their own liking 1186173657 M * daniel_hozac the kernel has a stable API. 1186173662 M * Bertl slack101: and as you should know, the kernel doesn't matter a bit 1186173672 M * slack101 well 1186173677 M * slack101 it depends on how you look at that 1186173688 M * Bertl slack101: otherwise you would not be able to run a debian guest on a fedora kernel, no? 1186173699 M * slack101 Bertl: depends on your app 1186173700 M * slack101 i guess 1186173720 M * slack101 i would use a full system vmware or qemu virtulization to test software on different distros though 1186173720 M * Bertl slack101: okay, name the app which depends on your distro kernel :) 1186173725 M * neuralis slack101: what are you talking about? 1186173747 M * daniel_hozac neuralis: i don't think you've been introduced to slack101. he's the virtualization king. 1186173748 M * Bertl neuralis: please, do not spend time on it :) 1186173764 M * slack101 im not going to argue theres been 2 times 1186173774 M * slack101 where vserver would not be right to test an app 1186173777 M * neuralis Bertl, daniel_hozac: got it. 1186173801 M * slack101 i look at vserver for isolation 1186173805 M * slack101 not for virtulization 1186173827 M * daniel_hozac and why would you need virtualization to test software? 1186173833 M * slack101 Bertl: for instance would could never install vbox on vserver 1186173850 M * daniel_hozac or kernels! yes you're right! 1186173863 M * daniel_hozac *cough* 1186173873 M * slack101 or anything to do with certain kinds of networking 1186173918 M * slack101 or an app which access certain hardware 1186173918 M * slack101 lol 1186173925 M * slack101 which is a quite big load of apps per say 1186173933 M * slack101 atleast the ones we were wroking with but anyways 1186173983 M * slack101 theres many instances where a vserver of any isloation based thing would not fit the bill 1186174016 M * Bertl daniel_hozac: hmm, you are wrong on that :) 1186174031 M * Bertl daniel_hozac: uml works quite nicely inside a guest :) 1186174035 M * daniel_hozac well, that's true. 1186174040 A * slack101 sighs 1186174072 M * Bertl slack101: I have the perfect project for you 1186174145 M * slack101 does it involve money ? 1186174149 M * Bertl slack101: you should run QEMU inside Linux-VServer inside Vitualbox on your system, and then simply put the Virtualbox inside QEMU, thus eliminating the hardware completely 1186174171 M * daniel_hozac lol 1186174190 Q * opuk Quit: leaving 1186174195 M * slack101 funny thing is my quadcore could most likely accomplish that in a very hardware idealistic manner 1186174274 M * slack101 anyways linux is linux every dsitro is 99.9999 percent the same with minor differences if your app wont run in debian its not going to run in red hat either 1186174300 M * Bertl you mean, every distro is 99.99% different :) 1186174310 M * daniel_hozac says the person who just said you need to run the distro kernel? 1186174319 M * slack101 well depends 1186174324 M * slack101 some distros cna bring excpetions 1186174326 M * slack101 can* 1186174328 M * Bertl on the argument? 1186174331 M * slack101 CentOS being one 1186174521 M * slack101 i could say the sky was blue you guys would say oh its grey 1186174535 M * Bertl currently it is black here :) 1186174553 M * slack101 lol 1186174561 M * Bertl slack101: but do you know _why_ the sky is blue and not e.g. green or red? 1186174576 M * slack101 water? 1186174605 M * Bertl if you are under water, maybe :) 1186174613 M * slack101 i remember reading stuff on it a while back how the sky is 4 shades blue 'er here then up borth 1186174616 M * slack101 north 1186174630 J * opuk ~kupo@213-64-112-143-no19.tbcn.telia.com 1186174704 M * slack101 but i have heard different answers from every person ...... same crap like the color of your blood 1186174715 M * slack101 its blue no its dark red 1186174731 M * Bertl slack101: look up rayleigh scattering in wikipedia 1186174811 M * slack101 i have actually read that before in a popular science magazine 1186174825 M * Bertl why am I not suprised :) 1186174844 M * slack101 but i have read about 5 other things aswell 1186174860 M * Bertl at the same time, I presume :) 1186174873 M * slack101 no, my brother had a subscription to PS 1186174874 M * slack101 but 1186174875 M * slack101 no 1186174902 M * slack101 i just hear different things all the time ;) 1186174935 M * Bertl that probably explains why you are saying different things all the time too :) 1186174945 M * slack101 lol 1186174961 M * slack101 man o man 1186174973 M * slack101 nope i just ramble to much 1186175007 M * slack101 Bertl: dont confuse rambling with incorrect information :P 1186175034 M * daniel_hozac you seem to be full of both. 1186175047 M * slack101 i love it 1186175054 M * slack101 you sit back just waiting 1186175081 M * Bertl daniel_hozac: do you know southpark? 1186175093 M * daniel_hozac know of it, haven't seen much of it though. 1186175162 M * slack101 Bertl: is this where you relate me to some character of it ? 1186175174 M * daniel_hozac slack101: or, i'm doing other things and whenever i look over here you're just telling some sort of lie :) 1186175187 M * Bertl daniel_hozac: a pity, this reminds me of the episode where Cartman 'tries' to kill the others with a plastic baseball bat 1186175211 M * slack101 those who watch and enjoy southpark are brain dead 1186175222 M * Bertl daniel_hozac: the punch line is 'I have to kill you ... it may take some time' 1186175274 M * Hollow oh, am i braindead 1186175276 M * Hollow :) 1186175280 M * daniel_hozac hehehe 1186175282 M * slack101 if you lik that show 1186175293 M * Hollow well, i saw every episode... in order 1186175294 M * Hollow :P 1186175301 M * daniel_hozac in one week? 1186175306 A * daniel_hozac loves a good marathon. 1186175306 M * Hollow no, not that fast 1186175313 M * Hollow but pretty fast :) 1186175333 M * slack101 only thing i watch sometiems that is half braindead ish is the simpsons 1186175347 M * slack101 and i saw that movie 1186175393 M * slack101 and that wasn't my idea 1186175398 M * Hollow still have to watch it .. there is a good german telesync :) 1186175420 M * Hollow but i guess they dubbed english audio meanwhile too 1186175425 M * daniel_hozac telesync and good don't belong in the same sentence :P 1186175439 M * Hollow well, it is rather good quality for TS 1186175452 M * Bertl well, and we are off topic :) 1186175455 M * Hollow hehe 1186175475 M * slack101 its not even a good mvoie but yea 1186175504 M * Hollow btw, mediawiki does not seem to support external redirects 1186175544 M * daniel_hozac do we need it? doesn't the installation on gentoo page already have everything? 1186175551 M * rgl Hollow, what is an external redirect? 1186175562 M * daniel_hozac redirect to an external page. 1186175566 M * Hollow rgl: a redirect outside the wiki 1186175579 M * Hollow we can just put a lonk to the gentoo howto in there... 1186175581 M * Hollow *link 1186175586 M * daniel_hozac Hollow: you've already done that ;) 1186175587 M * rgl didn't known that had a specific name hehe 1186175590 M * Bertl daniel_hozac: I would like to see something there as there is a special page for OVZ which is linked over and over again 1186175593 M * Hollow did i? 1186175624 M * rgl just make apache do the redirect for you? 1186175630 M * daniel_hozac Bertl: http://linux-vserver.org/Installation_on_Gentoo isn't good enough? 1186175630 M * Hollow oh, right.. i even copied the howto 1186175658 M * Hollow needs an update soon 1186176306 M * rgl gag... found the guilty options that crashes my system! 1186176324 M * daniel_hozac which is it? 1186176335 M * rgl its the framebuffer stuff 1186176514 Q * dna Quit: Verlassend 1186176629 M * rgl more exactly, its when I enable "Intel 830M/845G/852GM/855GM/865G/915G/945G support (EXPERIMENTAL)" I've replaced it wil VESA and it seems to work now. 1186176664 A * rgl has a 945GC chipset. 1186176680 M * daniel_hozac interesting... 1186176832 M * slack101 anyways i will make it my new irc resolution to not jabbr :P 1186177126 Q * FireEgl Quit: Bye... 1186177681 M * rgl OK finally got it working! yay :) 1186177697 M * rgl Linux rgl-desktop 2.6.22.1-rgl-desktop #14 SMP Fri Aug 3 22:45:41 WEST 2007 x86_64 *G* 1186177701 M * rgl 14 times *G* 1186178283 M * daniel_hozac my kernel is at 28 now i think :) 1186178337 M * Bertl #74 here :) 1186178376 M * rgl woah! you guys beat me by far *G* 1186178377 M * daniel_hozac my trees tend to get to a point where i don't remember what patches they have fairly quick... 1186178418 J * Aiken ~james@ppp121-45-195-95.lns1.bne1.internode.on.net 1186178800 M * Bertl wb Aiken! 1186178874 M * rgl humm. I've tried the new coretemp module, and it says the core temperature is at 30ºC, but lm-sensors say its at 46ºC ... :| 1186178891 M * Aiken good morning Bertl 1186178898 M * Bertl rgl: what does acpi report? 1186178937 M * rgl Bertl, humm, how can I see it with acpi? 1186178967 M * rgl /proc/acpi/thermal_zone/ is empty here. 1186179175 Q * bonbons Quit: Leaving 1186179761 J * meandtheshel1 ~markus@85.127.102.41 1186179915 M * rgl what happens when I choose the default of "Remap Source IP Address (VSERVER_REMAP_SADDR) [N/y/?] (NEW)"? the guest will use 127.0.0.1? 1186179976 M * Bertl rgl: it will not remap the source address :) 1186180041 M * rgl Bertl, that means multiple guests will have the same IP address, right? 1186180058 M * rgl Bertl, but with that enabled, its impossible for the guest to use 127.0.0.1? 1186180065 M * Bertl not unless you assign the same ip to more than one 1186180082 M * rgl 127.0.0.1 is assigned to more than one, no? 1186180091 M * Bertl usually not 1186180111 M * daniel_hozac usually it's not assigned at all :) 1186180125 M * rgl I mean, 127.0.0.1 is not available to all guests? or only the IP we have configured are available to the guest? 1186180410 M * rgl having a dual core is so sweet! it compiling in a core, and I'm doing stuff normally without any noticeable perf loss for *G* 1186180427 A * rgl is the first he using a dual core on desktop *G* 1186180535 M * rgl isnt' there a way to make the linux build run in paralel? 1186180547 M * daniel_hozac make -j2 1186180560 M * daniel_hozac (increasing 2 to whatever value you feel comfortable with) 1186180626 M * rgl cool. 2 is the right number I guess. one proc to each core. 1186180816 J * Piet ~piet@tor.noreply.org 1186181376 M * rgl Linux rgl-desktop 2.6.22.1-vs2.2.0.3-rc1-rgl-desktop #1 SMP Fri Aug 3 23:36:06 WEST 2007 x86_64 GNU/Linux :) 1186181730 M * rgl ok. I'm calling it a night! good night guys. have a nice one! 1186181741 M * Bertl u2 1186181742 Q * rgl Quit: Leaving 1186182059 M * matti Heh. 1186182106 M * daniel_hozac Bertl: btw, do you think this networking stuff will make it into 2.2? i.e. should it use a VCI-bit or should i just use version >= 0x00020300? 1186183059 M * Bertl hmm, good question 1186183112 M * Bertl I guess detecting that by the availability of the syscall command is a problem (otherwise you would not aks, yes?) 1186183131 M * Bertl *ask 1186183188 M * daniel_hozac well, currently the macros will only execute the best matching syscall. 1186183192 M * Bertl chances are good that it will get into 2.2.x sooner or later, so I think we should use VCI information for that 1186183204 M * daniel_hozac okay. 1186183429 M * Bertl I'll think about a name tomorrow :) 1186183452 M * Bertl something like NGNET2 or IPVX maybe .... 1186183467 J * noobasaur ~noobasaur@dsl-245-173-19.telkomadsl.co.za 1186183474 M * Bertl welcome noobasaur! 1186183485 M * noobasaur thank you :-) 1186183487 M * Bertl anyway, off to bed now ... have a good one everyone! 1186183502 N * Bertl Bertl_zZ 1186183505 M * daniel_hozac okay, good night! 1186183521 M * noobasaur night Bertl 1186183661 M * noobasaur I'm running a Debian Etch server in a colo which I manage remotely. I was trying to install linux-image-vserver-686 on the machine as per http://howtoforge.com/linux_vserver_debian_etch but after rebooting I can no longer ssh into the host. 1186183684 M * noobasaur is this at all related or have i been a n00b :) ? 1186183725 M * daniel_hozac what kernel was it running previously? 1186183729 M * daniel_hozac does the box come up at all? 1186183821 M * noobasaur hmmm .. you have me there ( on the previaous kernel) .. and I can't tell if the box is up or not. As luck would have it there were no other external service running on it, and I've configured firehol to ignore icmp messages - so either it's still down, or ssh isn't up. 1186183821 Q * ktwilight_ Read error: Connection reset by peer 1186183848 J * ktwilight_ ~ktwilight@91.80-66-87.adsl-dyn.isp.belgacom.be 1186183916 M * noobasaur I was curious as to how the vserver kernel install differed from a normal kernel install ... I've not done too many kernel install on live hosts before and wondered if there was something I'd overlooked? 1186184090 Q * Piet Quit: Piet 1186184844 M * daniel_hozac when you're using the package, it's supposed to do everything needed. 1186185001 M * noobasaur mm. I guess I assumed my networking would remain stable until after the reboot. I'm gonna see what's on the FAQ. 1186185097 M * neuralis compiling a kernel can't possibly have any effect on your networking. 1186185558 M * noobasaur yeah .. I'm not convinced that apt-get only compiles though ... 1186185581 M * daniel_hozac apt-get install doesn't compile at all.