1132790464 M * derjohn miller7, hello. anything particular ? I think there not much traffic 'ere today 1132790558 M * miller7 I can paste a 1500 pages manual if you want :P 1132790578 M * derjohn I have enough ebooks in-queue ;) 1132790589 M * derjohn (gonna habe to read ;)) 1132790627 M * derjohn I just asked because I'm abiut to leave - just wanted to know if there is a Q, before I leave thats all :) 1132790705 M * miller7 nah :) 1132790707 M * miller7 thank you though 1132790735 M * derjohn ok - you just wanted to HELO? OK - greetz back :) 1132790749 M * miller7 :) good night 1132790800 M * derjohn miller7, THX :) I just wait for my rc8 debian kernel got get finished - have to type my pass for signing, then I am gone ;) 1132790857 Q * ryker Quit: Leaving 1132790862 J * ryker ~ryker@c-67-176-240-74.hsd1.in.comcast.net 1132790935 Q * ryker Quit: 1132791206 J * brc bruce@201008046201.user.veloxzone.com.br 1132791209 M * brc Bertl alive ? 1132791220 M * miller7 I sure hope so 1132791226 M * miller7 he's too young still 1132791294 M * derjohn brc, Bertl said, he's back later. 1132791317 M * miller7 in his Terminator voice 1132791338 M * brc i am having a huge probelm 1132791342 M * derjohn miller7, the depends on the point of view - that would mean I am young too !! cool ! 1132791357 M * miller7 brc: can we help? 1132791383 M * miller7 derjohn: how old are you? 1132791411 M * derjohn miller7, 30 ... at least for this year ;) 1132791426 M * miller7 ok... there is someone I know who's older than that cough cough 1132791474 M * derjohn miller7, :) 1132791506 M * derjohn miller7, *lol* not I got the one with terminator voice 1132791511 M * derjohn not=now 1132791529 M * miller7 you need a CPU upgrade :P 1132791546 M * miller7 or renice some processes 1132791548 M * derjohn I habe an multibrain already 1132791567 M * derjohn Maybe only cooling ;) 1132791616 M * derjohn brc, your huge problem vanished? 1132792004 M * sizo n8 1132792011 M * miller7 bye sizo 1132792034 M * brc Just fixed. 1132792048 M * brc I was getting permissioned denied when trying to ping a vserver's ip 1132792056 M * brc Though it had something to do with permissions 1132792085 M * brc It was a firewall rule. maybe it is time to sleep :) 1132792100 M * derjohn brc, glad that we could help := 1132792101 M * derjohn ) 1132792271 M * miller7 sometimes even asking or being with people that don't know the problem helps 1132792277 M * miller7 like discussing with them and clearing your head 1132792283 M * miller7 then you find the solution :) 1132792303 M * derjohn miller7, yes, we should discuss our age ;) 1132792330 M * derjohn miller7, how many kernel patches did you submit in your life? 1132792330 M * miller7 eeeerhm 1132792337 M * miller7 none 1132792349 M * miller7 I was busy playing with girls 1132792368 M * miller7 oh well 1132792372 M * miller7 did neither of both 1132792372 M * derjohn miller7, hm, this is an fine alternative! 1132792375 M * miller7 I suuuuck 1132792377 M * miller7 :P 1132792406 M * derjohn miller7, I dont assume you mean that "by word" 1132792407 M * derjohn :) 1132792413 M * miller7 nope :P 1132792422 M * miller7 metaphorically speaking 1132792440 M * derjohn miller7, yes, in this channel obviously only geek hang out. 1132792445 M * miller7 your CPU works faster on such things, huh? :) 1132792470 M * derjohn miller7, errrmmmm ;) 1132792489 M * derjohn miller7, have to find a girl to play with ... in my age 1132792502 M * miller7 ough, that's old 1132792506 M * miller7 find a 20 year old one 1132792518 M * miller7 :) 1132792518 M * derjohn miller7, hm, yes, that might fit ! 1132793048 J * Aiken_ ~james@tooax7-230.dialup.optusnet.com.au 1132793370 Q * Aiken Ping timeout: 480 seconds 1132793491 J * ryker ~ryker@c-67-176-240-74.hsd1.in.comcast.net 1132794133 M * ag- vhashify power! 1132794138 M * ag- Program received signal SIGSEGV, Segmentation fault. 1132794138 M * ag- [Switching to Thread -1209736448 (LWP 17302)] 1132794138 M * ag- 0xb7faec5a in sha1Digest () from /usr/lib/libbeecrypt.so.2 1132794200 M * ag- many thanks to debian for bringing us such wonderful crappy shlibs ;) 1132794957 N * Bertl_oO Bertl 1132794960 M * Bertl back now ... 1132795311 M * TheSeer her bertl ;> 1132795316 M * TheSeer hey even 1132795330 M * TheSeer Nov 24 02:26:01 mx crond[19881]: System error 1132795336 M * TheSeer i seem not to be able to get rid of those.. 1132795341 M * TheSeer :-// 1132795364 M * brc that's true, in my case sleeing would help :) 1132795403 M * Bertl TheSeer: well, system error is very descriptive, when does crond output those? 1132795431 J * kai ~kai@221.137.212.203 1132795447 M * TheSeer Bertl: whenever an entry in the root-crontab is due 1132795447 M * Bertl welcome kai! 1132795478 M * Bertl TheSeer: well, yes, I assumed that, but we are interested in the source code of your crond 1132795587 M * TheSeer trying to get it ;> 1132796153 M * miller7 what is the CTX 1 vserver that runs at times???? 1132796159 M * TheSeer the root 1132796162 M * TheSeer aka host 1132796175 M * miller7 isn't it CTX 0? 1132796189 M * Bertl miller7: xid=1 is the spectator context 1132796204 M * miller7 what does it spectate? And who told it to spy on me? :P 1132796212 M * TheSeer oh yeah.. it's to late for me ;> 1132796213 M * Bertl miller7: it will see (and partially be able to interact) with all processes 1132796221 M * miller7 so as... ? 1132796234 M * Bertl miller7: try 'vps' 1132796250 M * miller7 I do all the time... does it use CTX 1 for it? 1132796263 M * Bertl sure, that's the only way to get all info 1132796267 M * miller7 aha 1132796270 M * miller7 didn't know :) 1132796274 M * miller7 was it like that in the past also? 1132796292 M * Bertl yes, I think so, it was like this for a long time now :) 1132796320 M * miller7 never noticed it :) 1132796328 M * miller7 I thought CTX 0 was responsible for that or somethign 1132796329 M * miller7 hehe 1132796332 M * miller7 oh well 1132796335 M * miller7 thank you beeeeeeeeert 1132796360 P * kai Leaving 1132796365 M * Bertl miller7: you're welcome! 1132796625 M * TheSeer hm.. i'll look into that tomorrow.. 1132796638 M * Bertl ok, np 1132796645 M * TheSeer *ywan* 1132796650 M * TheSeer i can hardly focus ;-P 1132796655 M * TheSeer okay.. heading home.. 1132796661 M * miller7 you can't hardly yawn properly :P 1132796668 M * Bertl lol 1132796672 M * TheSeer hmpf ;> 1132796681 M * TheSeer that's bad, isn't it? 1132796690 M * miller7 yep 1132796696 M * Bertl TheSeer: terrible! 1132796698 M * TheSeer :) 1132796703 Q * infowolfe Remote host closed the connection 1132796710 M * TheSeer okay.. off ;) 1132796715 M * miller7 take care TheSeer 1132796749 J * infowolfe ~infowolfe@209-112-215-110-cdsl-rb1.nwc.acsalaska.net 1132797325 M * anonymousc Bertl: where should I look for the latest docs on the COW Immutable Link Breaking in 2.1? 1132797356 M * Bertl hmm, not sure there is much docu on that ... 1132797379 M * anonymousc was there a particular day on which it was discussed via irc? 1132797400 M * Bertl check that out: http://www.13thfloor.at/vserver/search/ 1132797408 M * anonymousc already did :) 1132797432 M * Bertl k, haven't found anything useful? what do you want to know? 1132797516 M * anonymousc found some notes on people's experiences - was just looking for a description of exactly how it works so I can works out whether its applicable in my situation 1132797553 M * anonymousc obviously the most noticable thing is that it only works on filesystems that support the immutable flag 1132797587 M * Bertl yep, well, it's pretty simple, you've probably read how unification works, yes? 1132797602 M * anonymousc yeah - gotta love the simplicity of that 1132797634 M * Bertl good, now the CoW link breaking just adds an automatic 'copy that link if written to' feature 1132797664 M * anonymousc and that works for *all* types of writes? 1132797675 M * Bertl it uses the same basic mechanism, nothing new there, but instead of returning an error, it makes a copy 1132797688 J * bwana ~bwana@68-117-34-122.dhcp.roch.mn.charter.com 1132797693 M * bwana alo again 1132797696 M * Bertl welcome bwana! 1132797729 M * Bertl anonymousc: it is supposed to work on _all_ writes, although the semantics of read/write combinations are not precisely identical with 'normal' filesystems 1132797807 M * bwana i was combing through the vserver site and found a good article on install, but since then have lost it and can no longer find it. I do remember that this file had a suggestion about creating a hard link to the kernel source to make it easier to patch and such. Does anyone know to what article I am refering? 1132797840 M * Bertl bwana: hmm, let's ask google, shall we? 1132797848 M * anonymousc bwana: http://linux-vserver.org/Step-By-Step+Guide+2.6 1132797883 M * bwana ya thats it 1132797886 M * bwana thanks anonymousc 1132797927 M * bwana god how do i get to that from the main site 1132797985 M * anonymousc Bertl: so the cow writes get away from the previous reliance on the package manager doing the mv ; copy ; delete inside the vserver 1132798036 M * Bertl yes, Aiken tested 100% hard linked guests, i.e. even log files were linked 1132798124 M * Bertl bwana: well, it's on the documentation page :) 1132798208 M * bwana hmm 1132798247 M * Bertl first entry under Howtos 1132798248 M * bwana bah there it is 1132798256 M * bwana thanks 1132798260 N * Aiken_ Aiken 1132798270 M * Bertl bwana: you're welcome! 1132798284 M * Bertl morning Aiken! 1132798323 M * Aiken good lunch time :) 1132798344 M * Aiken I got nowhere with vroot as a module 1132798358 M * Bertl I fixed it yesterday, should work now 1132798443 Q * infowolfe Remote host closed the connection 1132798446 M * Bertl was a basically trivial issue 1132798475 M * Bertl forgot that CONFIG_* isn't define for modules and you have to use CONFIG_*_MODULE 1132798476 J * infowolfe ~infowolfe@209-112-215-110-cdsl-rb1.nwc.acsalaska.net 1132798569 M * Aiken just read Chuck's last message to the list, I have had gentoo do that in the past. Nasty 1132798683 M * bwana does the rc8 patch work well 1132798708 M * Bertl I hope so :) 1132798713 M * bwana heh 1132798723 M * miller7 Aiken: what are you referring to? 1132798736 M * Aiken the unmerging of glibc 1132798786 M * Aiken I have had the alpha running rc8 and a vserver in that machine using squid to act as a proxy today with no problems 1132798791 M * miller7 what's the title of this email? 1132799227 M * Aiken Re: [Vserver] gentoo guest wont start up or find anything after failed emerge 1132799925 M * anonymousc if a gentoo vserver is failing you can chroot into /vserver/ and maniupulate it, even if it won't start as a vserver (same with any guests actually) 1132799947 M * Bertl depends, if the libraries are broken, no luck :) 1132799958 M * miller7 the host libraries? 1132799966 M * Bertl no, the guest libraries 1132799990 M * miller7 well you can always chroot and re-emerge base packages I think 1132800025 M * Bertl chroot executes something in the changed root, so as soon as this requires one of the broken libraries, it will fail 1132800026 M * Aiken if the host is gentoo maybe something like ROOT=/vservers/whatever emerge glibc to replace the missing glibc 1132800056 M * miller7 well, untarring stage3 will make it work and then go from there again 1132800056 M * Bertl miller7: try it out, remove your libc inside a guest 1132800071 M * Aiken nasty 1132800216 M * miller7 Bertl: I think that extracting stage3 would solve that problem 1132800226 M * miller7 might try it and see 1132800410 M * anonymousc obvously if the libs inside the sytem are busted then you need to copy in working ones (or place a static shell in there and chroot and run to that). replace broken libs as required 1132800453 M * miller7 actually a gentoo guest is a nice way to see if gentoo upgrades can screw your dedicated boxes :) 1132800460 M * Bertl anonymousc: well, obviously not so obviously :) 1132800512 M * anonymousc in my experience most gentoo vservers wouldn't "start" because /var/lib/init.d/started was still populated. however that's not to say that your glibc isn't borked :) 1132800582 M * miller7 sometimes you need to zap the service in order to start 1132800586 M * miller7 which can be annoying 1132800631 M * anonymousc nope - you put "rm -f /vservers/corp/var/lib/init.d/started/*" into /etc/vservername>/corp/scripts/pre-start 1132800647 M * anonymousc hmm make that: /etc/vservername>/corp/scripts/pre-start 1132800681 M * anonymousc er - make that /etc/vservers//scripts/pre-start 1132800693 M * anonymousc cut-n-paste skillz letting me down 1132800782 M * miller7 LOL, some guest has 1.4GB in /var/spool/mail/root 1132801504 Q * Johnnie Quit: G'bye! 1132801645 M * anonymousc hmm - given that ocfs2 is built upon the ext2/3 disk format (with at least one 64bit change from memory) i wonder whether it would support running a vserver (as an alternative to using nfs) and also support immutable flags (possibly the cow links with the appropriate patching...) 1132801881 J * Johnnie ~john@acs-24-154-53-217.zoominternet.net 1132802041 M * Bertl anonymousc: well, try and let us know :) 1132802597 M * anonymousc given the rarity of spare san gear, its probably going to be done on top of drbd, gnbd, iscsi or aoe for testing (already managed to wedge an ocfs2 machine accessing disk via the aoe software (using vblade from aoetools for those interested). 1132802654 M * Bertl aoe, drbd and iscsi was already reported working :) 1132802740 M * Bertl (with ext2/3 IIRC) 1132802822 M * anonymousc is the majority of the immutable link code in the vfs or down in the actual filesystem implementation? 1132802860 M * Bertl all the checks and handling is at the vfs layer, the implementation is on the fs layer 1132802926 M * anonymousc mmm - off to linux-cluster to ask some questions... 1132804496 Q * ryker Ping timeout: 480 seconds 1132804573 J * ryker ~ryker@c-67-176-240-74.hsd1.in.comcast.net 1132804581 J * lilo_ ~lilo@lilo.usercloak.oftc.net 1132804852 Q * lilo Remote host closed the connection 1132805897 M * dlippolt what is the ONBOOT="yes" equivalent in the new management system? 1132805903 M * dlippolt not seeing it on the flower page 1132805918 M * dlippolt rather, in my local copy of the flower page :) 1132806180 Q * Aiken Ping timeout: 480 seconds 1132807302 M * Bertl dlippolt: the marks 1132808058 M * bwana what should i use for the memory split with 1 gig of ram, default ro maybe 2.5/1.5 1132808195 M * Bertl 2/2 should be the best, disable highmem 1132808286 M * bwana ok thnaks 1132808345 M * Bertl np 1132808371 M * anonymousc incidentally - with highmem 2/2 on a 4 gig box, each process can only address 2G itself, but can you run, say, 3 processes each using 1G and be utilizing the 3 of the 4 gigs of ram for userland processes? 1132808401 M * miller7 what kind of process needs that RAM? 1132808472 M * anonymousc thats just an example - in reality my issue is one of many, many processes each using 10-20 megs (can you guess what it is) 1132808516 M * anonymousc just wondering if its worth bringing some 32bit xeon boxes up to 4 gigs (from 2) whilst attempting to avoid the highmem issues 1132808572 M * Bertl anonymousc: the problem is different ... 1132808594 M * Bertl with 4GB memory, on x86, there is no way to avoid himem 1132808614 M * Bertl so, it doesn't really matter _what_ split you use 1132808645 M * Bertl on a 3GB box, you can use almost all the memory without himem 1132808653 M * anonymousc ah - so the 2/2 is purely an address *space* choice 1132808660 M * miller7 what's the benefit again of not using highmem? 1132808668 M * miller7 in practical terms I mean 1132808671 M * Bertl miller7: speed and performance 1132808678 M * anonymousc and i'll probably loose a chunk to pci address space between 3 and 4 anyway... 1132808687 M * miller7 how much better speed and performance? 1132808713 M * anonymousc bounce buffers and problems with certain caches which must live in lowmem 1132808754 M * Bertl I have no test results for that, but as the tlb stuff is still quite expensive, I'd say you might lose around 20% in unfortunate situations 1132808782 M * miller7 define unfortunate pleae 1132808884 M * Bertl heavy disk I/O and memory pressure? 1132808913 M * miller7 ic 1132808922 M * miller7 common problem 1132808983 M * anonymousc should i still use a 2/2 split with 3G of ram? 1132809389 M * miller7 17 guests without unification on a P4 with 2GB RAM is ok? 1132809396 M * miller7 or too little? 1132809412 M * miller7 cause I got more to add to this box if I want to :P 1132811107 M * Bertl should be fine, unless they are gaming servers :) 1132811115 M * miller7 nah 1132813264 Q * ryker Quit: Leaving 1132813450 Q * marl_mobile Ping timeout: 480 seconds 1132813932 Q * Johnnie Read error: Connection reset by peer 1132815228 N * lilo_ lilo 1132816531 Q * lonewolff Ping timeout: 480 seconds 1132816762 J * lonewolff ~lonewolff@host86-128-133-145.range86-128.btcentralplus.com 1132818026 J * Johnnie ~john@acs-24-154-53-217.zoominternet.net 1132818583 Q * Johnnie Ping timeout: 480 seconds 1132819150 J * Johnnie ~john@acs-24-154-53-217.zoominternet.net 1132819662 P * miller7 1132820175 J * lilo_ tor@lilo.usercloak.oftc.net 1132820591 Q * lilo Ping timeout: 480 seconds 1132820715 J * oliwel ~mail-at-o@host-62-245-151-178.customer.m-online.net 1132820725 M * Bertl welcome oliwel! 1132820756 M * oliwel hi bertl 1132820771 M * anonymousc Bertl: which filesystems does COW work with? 1132820804 M * Bertl basically all fs which support sendfile/page and iunlink flags 1132820935 M * anonymousc ah - i thought I saw somewhere that you had reused an ext2/3 field to store the relevant info, so it might be tied to that fs. thats good news. 1132821001 Q * shedi Quit: Leaving 1132822183 M * bwana question regarding the timer frequency, my default is set to 250 documentation says the default should be 1000 1132822188 M * bwana which default is correct 1132822206 M * Bertl for 2.6.13/14 default is 250, for older kernels 1000 1132822231 M * bwana ok thanks 1132822249 M * Bertl for linux-vserver it doesn't really matter what you choos 1132822251 M * Bertl +e 1132822265 M * Bertl you only have to 'consider' the HZ for the scheduler 1132822298 M * bwana what scheuler 1132822317 M * Bertl the kernel scheduler (also prio and hard in linux-vserver) 1132822344 M * bwana ok 1132822761 M * bwana hmm what is the deal with not being able to make a change with power management 1132822778 M * bwana also pci it seems 1132822801 M * bwana is this "always on" then 1132822841 M * Bertl there are certain 'states' which force in other modules/selections 1132822886 M * Bertl for example, if you enable a feature, which 'requires' bla, but does not depend on bla, then it has to force bla in 1132822928 M * bwana huh never noticed it before in the kernel 1132822933 M * bwana or is it part of your patch 1132822938 M * Bertl not sure what would force in power management though ... 1132822962 M * bwana yea 1132822970 M * Bertl the linux-vserver kernel patch forces in a single module, the capability system 1132822988 M * Bertl (as it is required for operation :) 1132823035 M * bwana bullshit, it'll work fine without it 1132823040 M * bwana heh 1132823271 M * harry Bertl: what /bin/sh does chroot use? the one of the host system, or the one on the guest? 1132823286 M * Bertl the one of the guest 1132823309 M * harry weird then... 1132823408 M * Bertl well, probably depends on the chroot implementation, if you use busybox, it might be that it simply executes sys_chroot() and restarts itself as shell 1132823413 M * harry bin/bash: /lib/libc.so.6: version `GLIBC_2.3' not found (required by /bin/bash) 1132823424 M * harry it's an exact copy of the original system 1132823449 M * harry why should it, after a rsync, suddenly need a GLIBC_2.3 instead of GLIBC_2.2 ? 1132823501 M * Bertl how do you get this info? 1132823554 M * harry vserver1:/vservers/tobiaas# chroot . 1132823554 M * harry /bin/bash: /lib/libc.so.6: version `GLIBC_2.3' not found (required by /bin/bash) 1132823557 M * harry vserver1:/vservers/tobiaas# strings /lib/libc.so.6 |grep GLIBC_2.3 1132823559 M * harry GLIBC_2.3 1132823562 M * harry GLIBC_2.3.2 1132823565 M * harry GLIBC_2.3.3 1132823567 M * harry GLIBC_2.3.4 1132823570 M * harry vserver1:/vservers/tobiaas# strings lib/libc.so.6 |grep GLIBC_2.3 1132823589 M * Bertl well, bash does tell you that glibc was not found? 1132823601 M * harry vserver1:/vservers/tobiaas# true 1132823610 M * harry hmm.. 1132823646 M * harry why does the bash inside the vserver need GLIBC_2.3 ??? its not there on the original system 1132823668 M * Bertl I don't know if and if why it needs it .. 1132823681 M * Bertl the question is, how do you conclude that it is required? 1132823710 M * harry since it wont do anything and says: /bin/bash: /lib/libc.so.6: version `GLIBC_2.3' not found (required by /bin/bash) 1132823748 M * Bertl so, you do: 'chroot /path/to/guest /bin/bash' yes? 1132823821 M * harry vserver1:/vservers/tobiaas/bin# ls -l bash 1132823821 M * harry -rwxr-xr-x 1 root root 625228 2004-12-19 17:53 bash 1132823821 M * harry vserver1:/vservers/tobiaas/bin# chroot /vservers/tobiaas/ /vservers/tobiaas/bin/bash 1132823824 M * harry chroot: cannot run command `/vservers/tobiaas/bin/bash': No such file or directory 1132823827 M * harry vserver1:/vservers/tobiaas/bin# chroot /vservers/tobiaas/ /bin/bash 1132823830 M * harry /bin/bash: /lib/libc.so.6: version `GLIBC_2.3' not found (required by /bin/bash) 1132823866 J * marl_mobile ~matt@84.92.193.225 1132823866 M * Bertl well, the first one is simple, it's not there 1132823879 M * harry i wonder how it can NEED glibc_2.3 if it isn't there normally 1132823901 M * harry on the real system, there is no GLIBC_2.3 either, but everything works there :( 1132823920 M * Bertl maybe you did run ldconfig over the guest libs? 1132823952 M * harry hmm... not that i know of... 1132823961 M * harry who should i be able to fix it then? 1132823990 M * Bertl what does: strings /vservers/tobiaas/bin/bash | grep libc 1132823995 M * Bertl give? 1132824001 M * harry vserver1:~# strings /vservers/tobiaas/bin/bash |grep GLIBC 1132824001 M * harry GLIBC_2.1 1132824001 M * harry GLIBC_2.0 1132824001 M * harry GLIBC_2.3 1132824003 M * harry GLIBC_2.2 1132824028 M * Bertl so it seems your bash has different libs coded in 1132824045 M * harry can this be because of 2.4-2.6 kernel difference? 1132824056 M * Bertl and it's not unexpected that it 'mentiones' the last one, if none of the libraries can be found 1132824087 M * harry GOD FUCKING DAMMIT! 1132824091 M * Bertl what does chroot /vservers/tobiaas/ /bin/true give? 1132824094 M * harry it's solved, sorry for the inconvenience 1132824116 M * harry my collegue said he copied the bin directory again... 1132824120 M * harry but apparently, he didn't 1132824134 M * harry i resynced, and now it works... /me now feeling bad for waisting your time :s 1132824137 M * harry so... sry 1132824145 M * Bertl np, have fun! 1132824275 M * Bertl okay, off to bed now ... back later :) 1132824281 N * Bertl Bertl_zZ 1132824838 J * shedi ~siggi@tolvudeild-203.lhi.is 1132825441 Q * tanjix Ping timeout: 480 seconds 1132827051 M * harry how comes that a vps starts in runlevel 3 instead of 2 ??? 1132827076 M * harry vserver1:/usr/local/vserver/etc/vservers/frodo# grep default /vservers/frodo/etc/inittab 1132827079 M * harry # The default runlevel. 1132827082 M * harry id:2:initdefault: 1132827083 M * harry so it should start in runlevel 2! 1132827092 M * daniel_hozac are you using the plain initstyle? 1132827115 M * harry it's installed as a normal debian ... 1132827138 M * harry so.. i guess i am :) 1132827154 M * daniel_hozac so /etc/vservers//apps/init/style contains plain? 1132827159 M * daniel_hozac and not sysv? 1132827164 M * daniel_hozac (or is non-existant) 1132827206 M * harry vserver1:/usr/local/vserver/etc/vservers/frodo/apps/init# ls -l 1132827206 M * harry total 0 1132827210 M * harry inexistant 1132827238 M * daniel_hozac so, it should be starting in runlevel 3. 1132827277 M * daniel_hozac since you're not running an init in the guest, nor have a runlevel file in apps/init with 2. 1132827290 M * harry so... i should do: echo 2 > /usr/local/vserver/etc/vservers/frodo/apps/init/runlevel.start 1132827293 M * harry ? 1132827307 M * daniel_hozac sure. 1132827850 Q * monrad Quit: Leaving 1132828382 Q * oliwel Ping timeout: 480 seconds 1132830447 J * oliwel ~mail-at-o@host-62-245-151-178.customer.m-online.net 1132830556 Q * virtuoso Ping timeout: 480 seconds 1132830558 J * virtuoso ~s0t0na@shisha.spb.ru 1132830799 Q * brc Remote host closed the connection 1132831026 M * anonymousc argh - how the heck do you "make sure the guest has _all_ files marked as unified"? I'm playing with 2.1 and want to test out COW links. 1132831662 J * mcp ~hightower@wolk-project.de 1132832033 M * anonymousc hmm - kernel panic with xfs - lets give ext3 a go instead... 1132833008 J * NikDaPhreak ~NikDaPhre@193.24.241.34 1132833014 M * NikDaPhreak hi all 1132833040 M * anonymousc hi phreak 1132834033 M * ag- anonymousc: could you be more precise about the kernel panic? info, debugging, etc... 1132834300 M * anonymousc i was going to post it tomorrow when bertl comes back - but if you want to help - here's the info: 1132834339 M * anonymousc kjournald starting. Commit interval 5 seconds 1132834339 M * anonymousc EXT3 FS on hdb, internal journal 1132834339 M * anonymousc EXT3-fs: mounted filesystem with ordered data mode. 1132834339 M * anonymousc Unable to handle kernel NULL pointer dereference at virtual address 00000002 1132834339 M * anonymousc printing eip: 1132834341 M * anonymousc c0178cb1 1132834341 M * anonymousc *pde = 00000000 1132834343 M * anonymousc Oops: 0000 [#1] 1132834343 M * anonymousc PREEMPT 1132834345 M * anonymousc Modules linked in: xfs exportfs dm_mod sd_mod dpt_i2o scsi_mod eepro100 mii 1132834345 M * anonymousc CPU: 0 1132834347 M * anonymousc EIP: 0060:[] Not tainted VLI 1132834347 M * anonymousc EFLAGS: 00010282 (2.6.14.2-vs2.1.0-rc8) 1132834349 M * anonymousc EIP is at cow_break_link+0x1b1/0x2d0 1132834349 M * anonymousc eax: ca553ae0 ebx: cffe3c00 ecx: 00000001 edx: fffffffa 1132834351 M * anonymousc esi: c79efff5 edi: c5b07bb8 ebp: ccb14e58 esp: ccb14e20 1132834351 M * anonymousc ds: 007b es: 007b ss: 0068 1132834353 M * anonymousc Process rcS (pid: 23840, threadinfo=ccb14000 task=cf626590) 1132834353 M * anonymousc Stack: c5b07bb8 cffe3c00 00000001 ccb14e58 ccb14e4c c0153535 00000008 c79ef000 1132834355 M * anonymousc ca553ae0 fffffffa cffe3c00 c5b03dd8 000021b6 ffffffe1 c2d77c40 cffe3c00 1132834355 M * anonymousc 01ca14b9 00000004 c79efffa 00000300 00000000 00000000 cbeae5e0 c79f7000 1132834357 M * anonymousc Call Trace: 1132834357 M * anonymousc [] unmap_page_range+0x95/0xc0 1132834359 M * anonymousc [] strncpy_from_user+0x41/0x70 1132834359 M * anonymousc [] get_unused_fd+0xc2/0x100 1132834361 M * anonymousc [] path_lookup+0x8c/0x160 1132834361 M * anonymousc [] open_namei+0xdb/0x710 1132834363 M * anonymousc [] filp_open+0x58/0xa0 1132834363 M * anonymousc [] strncpy_from_user+0x41/0x70 1132834365 M * anonymousc [] get_unused_fd+0xc2/0x100 1132834365 M * anonymousc [] do_sys_open+0x4f/0xe0 1132834365 M * TheSeer moah 1132834367 M * anonymousc [] syscall_call+0x7/0xb 1132834367 M * anonymousc Code: db 74 03 ff 43 28 89 5c 24 04 b8 01 00 00 00 89 44 24 08 89 3c 24 e8 1f b4 fe ff 89 44 24 20 85 c0 0f 84 a3 00 00 00 8b 54 24 24 <8b> 42 08 8b 70 08 b8 01 00 00 00 e8 ef c7 f9 ff 8b 5e 40 b8 01 1132834386 M * TheSeer anonymousc: please.. paste something like that to a paste site 1132834410 M * anonymousc no kidding - sorry all 1132834449 M * anonymousc care to recommend one? 1132834474 M * TheSeer ouhm 1132834477 M * TheSeer phpfi.com 1132834484 M * TheSeer pastebin.com 1132834603 J * yungyuc ~yungyuc@220-135-53-220.HINET-IP.hinet.net 1132834830 M * anonymousc http://pastebin.com/437023 1132835043 M * anonymousc actually the panic happens when you try to start the clone or the original 1132835154 M * anonymousc *sigh* now its only the clone that causes the panic (makes sense i suppose since the original doesn't care about breaking the link) 1132835358 M * ag- anonymousc: echo 1 >/proc/sys/vserver/debug_misc might give us more info of what happened in cow_break_link() 1132835744 J * menomc ~amery@200.75.27.121 1132835851 Q * mnemoc Ping timeout: 480 seconds 1132835851 N * menomc mnemoc 1132836639 Q * oliwel Quit: Chatzilla 0.9.68.5 [SUSE 1.0.6-16/20050715] 1132837702 M * daniel_hozac anonymousc: addr2line -e /vmlinux c0178cb1 1132839002 J * monrad ~monrad@shop3.diku.dk 1132840946 Q * monrad Quit: Leaving 1132844547 J * vrwttnmtu ~eryktyktu@82-69-161-137.dsl.in-addr.zen.co.uk 1132845742 J * Larsioa ~lars@you.are.the.lamer.la 1132845744 M * Larsioa Helllooo ;) 1132846329 J * monrad ~monrad@213083190130.sonofon.dk 1132846861 Q * yungyuc Quit: leaving 1132847253 M * Larsioa derjohn: finally i got my sun cobalt raq 550 :D 1132847777 Q * Hollow Remote host closed the connection 1132848675 J * Hollow ~hollow@home.xnull.de 1132849518 N * Bertl_zZ Bertl 1132849522 M * Bertl morning folks! 1132849575 Q * shedi Quit: Leaving 1132849600 M * Bertl anonymousc: do you have the kernel tree at hand? 1132849932 M * vrwttnmtu Hello Bertl 1132850089 M * Bertl welcome vrwttnmtu! 1132850254 J * oliwel ~mail-at-o@host-62-245-151-178.customer.m-online.net 1132850265 M * Larsioa hi Bertl 1132850674 M * marl_mobile hi, ive got vserver running on a debian hosts system and can install debian based clients easly using method bootstrap, how do i install say a fc4 based client? 1132850704 M * Bertl hi Larsioa! marl_mobile! 1132850711 M * marl_mobile hi Bertl 1132850821 M * marl_mobile any pointers for the fc4 install? 1132850860 M * Larsioa use the deply-server script 1132850863 M * Larsioa read the faq 1132851270 M * Bertl marl_mobile: the problem on debian is, that it doesn't support rpm very well 1132851274 M * marl_mobile ok Larsioa ive already had a look at the faq, couldnt see anything before and cant find any references to deply-server script, am i looking at the write faq? http://linux-vserver.org/Linux-Vserver+FAQ 1132851300 M * Bertl marl_mobile: so, given that you manage to compile rpm, apt-rpm, or yum, you can simply use those isntallation methods 1132851344 M * marl_mobile so id need to compile apt-rpm AND rpm on the debian system, then the rpm-apt method should work? 1132851389 M * Bertl yes, definitely, but I have not heard from anybody who managed to do that 1132851420 M * marl_mobile it was just that under /etc/vservers/.distributions there are folders for fc1/2/3/4 with apt listings for them, thought it might have done it some other way 1132851428 M * marl_mobile thanks :( 1132851436 M * Bertl yes, that's for the apt-rpm 1132851455 M * Bertl which is available on many distros (even non rpm based) 1132851471 M * marl_mobile ok, ill fire up another debian client and try compiling apt-rpm and rpm :( ill let you know how i get on 1132851487 M * Bertl of course, you can always create a few guests on an rpm based machine, and copy them over ... 1132851541 M * Larsioa Has anybody in here ever worked with Sun cobalt raq 550 ? 1132851560 M * marl_mobile is there an easy way to do that from a standard installed machine? e.g. take an already running standard RH box, and copy trhe files directly into a vserver? 1132851621 M * marl_mobile is there any docs that point out which files to avoid copying? like do i need any proc entries? what about dev entries? 1132851691 M * Bertl yes, there is a howto for that on linux-vserver.org 1132851704 M * Bertl basically you do not copy /proc, as it is a virtual filesystem 1132851727 M * Bertl and you should replace the /dev with a newly created (secure) /dev of the skeleton build method ... 1132851766 M * Bertl it basically comes down to creating a skeleton with 'vserver build -m skeleton --context 42 --interface ... 1132851791 M * Bertl (see the alpha-util-vserver page for details and examples) 1132851802 M * Bertl then copy over everything except /dev and /proc 1132851829 M * Bertl after the first startup, cleanup packages you do not need anylonger and hardware related stuff 1132851852 M * marl_mobile thanks :) 1132851859 M * Bertl you're welcome! 1132852027 M * marl_mobile whats the address for alpha-utils-vserver i cant see it on the site :( 1132852115 M * baggins Bertl: do you want a patch for util-vserver that adds CPUSET support? 1132852157 M * baggins Bertl: I have a simple version now (just puts a vserver guest into cpuset) but I can extend it easily 1132852264 M * Bertl baggins: sure, against 0.30.209? 1132852302 M * Bertl baggins: in any case, please send it to Enrico and the ML with some comments on what it does and how it was tested 1132852319 M * baggins Bertl: ok, I'll do that 1132852348 M * Bertl CPUset support is definitely something we want in the future! 1132852351 M * baggins I tested it on quad dual-opteron system and it looks good 1132852422 M * baggins I'll add a create-on-build/start feature and some docs and send it later today or tomorrow 1132852475 A * baggins have to go now 1132852766 M * Bertl k, cya! 1132852848 M * marl_mobile going away for a LUG meeting speek to u later thanks again 1132852855 M * marl_mobile and greate work on vserver btw :) 1132853020 Q * marl_mobile Quit: Leaving 1132853273 J * shedi ~siggi@inferno.lhi.is 1132853581 Q * lilo_ Ping timeout: 480 seconds 1132853609 M * Bertl okay, off for now .. back later 1132853624 N * Bertl Bertl_oO 1132853832 J * lilo tor@lilo.usercloak.oftc.net 1132854321 Q * oliwel Quit: Chatzilla 0.9.68.5 [SUSE 1.0.6-16/20050715] 1132854928 J * stefani ~stefani@c-24-19-46-211.hsd1.wa.comcast.net 1132854987 P * stefani 1132855455 Q * Larsioa Ping timeout: 480 seconds 1132856722 J * sc2004 ~bkvpr@host81-134-88-88.in-addr.btopenworld.com 1132856992 M * dlippolt anyone have multicast working on vs2.0? 1132857205 Q * Hollow Read error: Connection reset by peer 1132858445 Q * Johnnie Ping timeout: 480 seconds 1132859093 N * Bertl_oO Bertl 1132859142 M * Bertl dlippolt: hmm, IIRC somebody was doing that a few days back ... is multicasting from vserver guests a new hobby? 1132859233 M * Megabart hello boy! 1132859263 M * Bertl hmm, which one? 1132859571 M * dlippolt Bertl: haha, sort of. attempting to buid tomcat clusters in vserver 1132859586 M * dlippolt Bertl: i was probably that person 1132859608 M * dlippolt Bertl: i have the new alpha tools up 1132859626 M * dlippolt Bertl: now i'm searching for the simplest multicat test-tool so we can browse strace's together :) 1132859706 M * Bertl ah, cool! 1132859729 M * Bertl somebody (please check the irc logs) provided simple test code, IIRC 1132859729 M * dlippolt Bertl: i'll be offline for a couple hours for turkey, but will pick this back up again this afternoon 1132859746 M * dlippolt ok,i'll check 1132859750 M * Bertl yeah, will probably be offline too for some time ... 1132859768 M * Bertl going to fly to Princeton tomorrow ... 1132859965 M * TheSeer wow.. 1132859976 M * TheSeer going to study there? ;) 1132859990 M * TheSeer or even teach? 1132860368 M * Bertl well, it's a cooperation regarding linux-vserver 1132860394 M * Bertl we will attack ngnet and similar bleeding edge stuff 1132860514 M * TheSeer nice ;) 1132860531 M * TheSeer how does the XEN stuff affect all this btw? 1132860546 M * Bertl xen is complementary to linux-vserver 1132860572 M * Bertl it is virtualization at the hardware level, where we do isolation (and some virtualization) at the application level 1132860584 M * Bertl xen is similar to UML just faster 1132860677 M * TheSeer hmm.. so no code sharing or stuff i guess.. 1132860703 M * Bertl not really, xen uses a hypervisor, and a moderately adapted virtual architecture (for the kernel) 1132860721 M * TheSeer k :) 1132860746 M * Bertl what we try (maybe xen too) is to keep them 'compatible' 1132860763 M * Bertl so that you can mix them easily (UML, XEN, linux-vserver) 1132861102 M * TheSeer nice ;) 1132861151 M * TheSeer Nov 24 20:44:01 mx crond[3149]: System error 1132861153 M * TheSeer *sigh* 1132861368 M * Bertl what do you get in the crond logs? 1132861570 Q * shedi Quit: Leaving 1132861652 M * TheSeer exactly that line i pasted 1132861660 M * TheSeer nothing more 1132861662 M * Bertl in all three logs? 1132861669 M * TheSeer three? 1132861680 M * Bertl usually crond has an info/warn and error log 1132861696 M * Bertl /var/log/cron/ 1132861730 M * TheSeer in fedora it's just one file /var/log/cron 1132861772 M * Bertl well, maybe other logs too? 1132861803 M * TheSeer ==> messages <== 1132861803 M * TheSeer Nov 24 20:44:01 mx crond(pam_unix)[3149]: session closed for user root 1132861803 M * TheSeer Nov 24 20:46:01 mx crond: System error 1132861803 M * TheSeer Nov 24 20:46:01 mx crond(pam_unix)[3536]: session closed for user root 1132861863 M * Bertl well, as I said yesterday, you will have to get the source, and look for the 'descriptive' message 'system error' :) 1132861890 M * TheSeer i grep'd through it already.. nothing found 1132861907 M * Bertl my crond tells long stories about what failed when and why ... but maybe on FC it doesn't ... 1132861939 M * Bertl maybe enable debugging (check manual pages) 1132861956 J * Hollow ~hollow@home.xnull.de 1132861964 M * Bertl welcome Hollow! 1132861966 M * TheSeer what cron do you use? 1132861972 M * Hollow hey Bertl 1132861976 M * TheSeer fedora comes with vixie-cron 1132861999 M * Bertl yep, same here, older versions though 1132862009 M * Bertl vixie-cron-3.0.1 1132862022 M * TheSeer hmm.. k.. 1132862025 M * TheSeer i'll check back from home.. 1132862027 M * TheSeer bbiab ;) 1132862235 M * harry pfffft... reading the entire vserver patch is a lot of work... 1132862275 M * harry are there any docs on the patches? what's done why? 1132862292 M * harry i understand most of it... but some docs would make it easier :) 1132862327 M * Bertl well, if you want to understand it, it's easier to look at the split versions 1132862344 M * harry split versions? 1132862358 M * Bertl and of course, the Linux-VServer Paper is a good help to understand certain parts 1132862364 A * harry wants to understand/modify/make better ;) 1132862373 M * harry where possible) 1132862376 M * Bertl http://vserver.13thfloor.at/Experimental/split-2.6.12.3-vs2.0/ 1132862388 M * Bertl will update the split for 2.0.1 pretty soon 1132862410 M * Bertl it's a breakdown of the patch according to categories 1132862417 M * harry interesting ;) 1132862425 M * Bertl IMHO they are pretty much self explanatory 1132862430 M * harry uhu 1132862436 M * Bertl YMMV 1132862441 M * harry the include thing is pretty logical (first one ;)) 1132862745 M * harry -#define PERCPU_ENOUGH_ROOM 32768 1132862746 M * harry +#define PERCPU_ENOUGH_ROOM 65536 1132862752 M * harry why double it? 1132862768 M * Bertl because linux vserver uses percpu space, and it was pretty low 1132862779 M * Bertl (i.e. certain configs would overflow) 1132862789 M * harry as in, not big enough, or just a precaution? 1132862803 M * Bertl not big enough if you e.g. use xfs :) 1132862843 M * harry is it big enough if you use reiser? 1132862854 M * Bertl maybe, didn't try ... 1132862854 M * harry and what's the memory/performance impact on that value? 1132862876 M * harry (/me just amateur kernel hacker ;))) 1132862876 M * Bertl the performance impact is basically zero, and the memory impact is 32k? 1132863165 M * harry return sprintf(buffer, 1132863171 M * harry how large is that buffer (proc.c) 1132863249 M * Bertl where is that line? 1132863308 M * harry + * linux/kernel/vserver/proc.c 1132863315 M * harry everywhere :) 1132863333 M * Bertl :) 1132863338 M * harry static int proc_virtual_info(int vid, char *buffer) eg 1132863363 M * harry (i want to see if it's overflowable somewhere ;)) 1132863369 M * harry probably not, but it never hurts to check:) 1132863384 M * Bertl the proc system started with a buffer of one page (memory) 1132863414 M * harry so, on i386 it's 4k 1132863468 M * Bertl it was recently changed to support so called sequencers 1132863503 M * Bertl would be nice to update the procfs stuff to them 1132863517 M * harry what are they? 1132863524 M * Bertl (i.e. something you could attack if you like to :) 1132863524 M * harry (sry, /me never heard of them) 1132863556 M * harry i don't want to do it, but it would be a shame to get in the host system by exploiting a proc bug ;0 1132863559 M * harry ;) 1132863560 M * harry (IF possible) 1132863577 M * harry seems kinda hard, since no strings are printed to them... but maybe smth else 1132863586 Q * sc2004 Quit: »¡« Scøøp Script 2004 »!« with Self clearing virus System! Download it at www.scoopsite.com 1132863587 M * Bertl not really, the info is of known length, so no way to get over the limit 1132863634 M * Bertl harry: search for seq_printf, you should find a few usages 1132863655 M * harry + for (i=0; inbipv4; i++) { 1132863656 M * harry + length += sprintf(buffer + length, 1132863662 M * harry +int proc_nid_info (int vid, char *buffer) 1132863666 M * harry in that function 1132863684 M * Bertl yup? 1132863698 M * Bertl nbipv4 can be max 16 1132863708 M * harry uhu 1132863709 M * harry :) 1132863711 M * harry i know... 1132863837 M * Bertl but keep scanning (and reporting) I'm glad somebody is checking the code from the 'auditing' perspective 1132863865 A * harry was just thinking... what if sprintf returns -1 1132863882 M * harry (btw, i'm sure andrewg auditted the code 1132863903 M * harry so he'll have found it if there are errors ;) 1132863911 M * harry it's just for fun and understanding :) 1132863959 M * Bertl I'd say if sprintf returns -1 we have a problem with the gcc :) 1132863970 M * harry hehe 1132863987 M * Bertl (at least I do not see any indication that the kernel version would ever return negative values) 1132863998 M * harry (i'm more used to looking at application code, where sprintf can easily return -1 ;)) 1132864018 M * Bertl np 1132864090 A * harry noticed a lot of code that takes the return value of snprintf or sprintf as marker to put a nullbyte at the end of a buffer 1132864104 M * harry or just use that in a write further on (which causes a infoleak 1132864119 M * Bertl in what way? 1132864136 M * harry from manpage: 1132864137 M * harry If the output was trun- 1132864138 M * harry cated due to this limit then the return value is the number of characters (not including the trailing '\0') which would have been written 1132864141 M * harry to the final string if enough space had been available. 1132864178 J * Breaker_uk ~bkvpr@host81-134-88-88.in-addr.btopenworld.com 1132864179 M * harry so, buf[1024]; len = snprintf (buf, 1024, argv[1]); buf[len] = '\0' 1132864182 M * harry that kind of stuff 1132864189 M * Bertl you are looking at userspace snprintf, right? 1132864194 M * harry yups 1132864200 M * Bertl okay, good :) 1132864203 M * harry i don't think snprintf exists in kernelspace :) 1132864208 M * Bertl it does 1132864212 M * Bertl vsnprintf 1132864214 M * harry but that's a very common mistake in code... 1132864220 M * harry same for that one, yes 1132864258 M * Bertl yeah, well, strings and character arrays always cause fun in userspace 1132864262 M * Bertl (especiallly for new coders) 1132864267 M * harry :) 1132864280 M * harry check rosiello.net in that case ;)) 1132864289 M * harry if you want exploiting fun :) 1132864324 M * harry anyway... back to the diffs :) 1132864451 M * harry in IPv6, you're gonna have a problem with the default buffersize i think ;) 1132864640 M * Bertl yup, that's why ipv6 is not supported yet :) 1132864651 M * harry lol 1132864660 M * harry you can't be serious :) 1132864715 M * vrwttnmtu Is harry the IPv6 man? 1132864717 M * harry is that vx_name[65] part used anywhere??? 1132864727 M * harry vrwttnmtu: /me, ipv6-man? 1132864733 A * vrwttnmtu kowtows to harry 1132864734 M * harry nonono 1132864755 M * vrwttnmtu Our father, who bringeth ipv6.... harry be thy name... 1132864762 M * harry hehe 1132864768 A * vrwttnmtu will be silent again 1132864775 M * harry if only the proc entry is the problem 1132864782 M * harry i could try fixing that ;) 1132864794 M * harry but before i do that, i have some work on the rest of the implementation 1132864806 M * harry and "porting" the ipv4 part to ipv6 :) 1132864856 M * FaUl whoooo ipv6-enabled vserver? with ngnet or with the old-style? 1132864931 M * harry weird... the vx_info struct contains a vx_name field, but it's never used, not in the kernel, not in userspace utils 1132865412 M * Bertl hmm, in my kernel source it is used :) 1132865431 M * Bertl case VHIN_CONTEXT: 1132865431 M * Bertl return vxi->vx_name; 1132865471 M * harry http://pastebin.com/437442 1132865492 M * Bertl but to clarify, it is there for userspace to store a name to each context .. basicall an empty char field to store arbitrary reference data 1132865504 M * harry uhu 1132865522 M * harry so how can i put data in it? (preferrably more than 65 bytes ;)) 1132865529 M * harry well... 65 might be enough too ;) 1132865559 M * Bertl not at all, the limit is 65 bytes, as for the other utsnames 1132865610 M * Bertl it's used in kernel/vserver/cvirt.c 1132865625 M * harry i see it now... /me tries to find why it's not in what i pasted 1132865626 M * Bertl vx_vhi_name(() 1132865638 M * harry lol 1132865644 M * Bertl because you greped in the includes only :) 1132865646 M * harry i was in /include 1132865686 M * Bertl okay, have to leave now (and get the luggage to the airport) will be back later .. I think :) 1132865703 N * Bertl Bertl_oO 1132865706 M * FaUl Bertl_oO: good luck 1132865709 M * harry cya 1132865711 M * harry and hf 1132865715 M * harry (and thx ;)) 1132865731 M * FaUl harry: ipv6 with the old or the new networking? 1132865740 M * harry FaUl: what about that? 1132865752 M * harry 21:34 < harry> in IPv6, you're gonna have a problem with the default buffersize i think ;) 1132865755 M * harry 21:37 < Bertl> yup, that's why ipv6 is not supported yet :) 1132865762 M * harry i was just thinking about the proc stuff 1132865774 M * FaUl harry: did I understand right that you are plannigng to implement ipv6-support in linux-vserver 1132865786 M * harry if i'm gonna try implementing it, i would first like to know why it isn't done yet 1132865790 M * harry what the problems are etc.. 1132865803 M * harry if i konw how, i will try my best 1132865807 M * harry know 1132865884 M * FaUl harry: maybe you'll better help with the new networking stack 1132865899 M * FaUl because it will - if i understand it right - support ipv4 as well as ipv6 1132865927 A * FaUl personally would like an ipv6-enabled vserver-kernel definitly 1132865973 M * FaUl but I think it would be better if you help with the new stuff instead of using energy and time in something which will be replaced soon 1132865980 M * FaUl just my opinion 1132866019 M * harry it will all depend on my boss's decission 1132866032 M * harry wether we'll use vserver or xen or vmware for virtualising 1132866051 M * FaUl make him to deside for vserver, they are realy cool :-) 1132866060 J * Johnnie ~john@acs-24-154-53-217.zoominternet.net 1132866086 M * harry i mainly have debian servers (no, i can't choose), other part has redhat servers (no problem there, except maybe kernel policies), and then we have the boss himself, who now has vmware and windowsmachines 1132866098 M * harry FaUl: i have a vserver running for testing 1132866116 M * harry on a dual xeon 3GHz with 1 gig ram and 2 146 gig scsi discs in raid1 1132866123 M * harry all lvm'ed and blabla... 1132866124 M * FaUl nice 1132866137 M * harry so i hope that can convince them to keep vserver 1132866147 A * FaUl has vserever on a dual-PIII-500 with 6 scsi-discs and 2 ide-discs at home :-) 1132866159 M * harry only thing i fear... we can't have all 3 virtualisation tools 1132866200 M * harry ifaik, xen needs a patched kernel to run their VM's 1132866202 M * FaUl but xen/vmware cost lot of system-resources. 1132866209 M * harry so that 's an advantage for vserver 1132866211 M * FaUl yes, thats another problem 1132866218 M * harry ahmm.... that's no problem 1132866221 M * FaUl but vserver needs a patched kernel as well :-) 1132866230 M * harry that''s my problem... if we're gonna choose for vserver 1132866241 M * harry we'll also get a quad opteron with 8 gig's of ram or so 1132866248 M * harry (or maybe 2 or 3...) 1132866279 M * FaUl ok, if thats no problem, then maybe vmware is what you want - or maybe qemu :-) 1132866287 M * harry advantage for xen: it's devellopped by a university, so they aren't that likely to ever quit develloping 1132866312 M * FaUl well, i guess bertl won't stop soon either 1132866322 M * harry vserver doesn't have a lot of sponsoring i think 1132866332 M * harry at least not comparable to a university 1132866347 M * harry FaUl: i hope not ;) 1132866358 M * harry but the number of devellopers is a lot smaller than xen 1132866359 M * FaUl well that maybe true 1132866375 M * harry so... i'll have to wait on what the boss decides 1132866386 M * harry i hope vserver, and i hope i can help then... 1132866387 M * harry ;) 1132866394 M * FaUl :-) 1132866397 A * FaUl either does 1132866590 Q * Johnnie Ping timeout: 480 seconds 1132866742 J * SeerHome ~theseer@c231204.adsl.hansenet.de 1132866967 Q * PerlOffice Read error: Operation timed out 1132867155 J * Johnnie ~john@acs-24-154-53-217.zoominternet.net 1132867652 J * PerlOffice ~stuart@220-253-85-202.QLD.netspace.net.au 1132869176 Q * NikDaPhreak Ping timeout: 480 seconds 1132869958 Q * Breaker_uk Quit: »¡« Scøøp Script 2004 »!« download it at www.scoopsite.com 1132870975 J * Aiken ~james@tooax8-133.dialup.optusnet.com.au 1132871151 J * yarihm ~yarihm@84-73-119-83.dclient.hispeed.ch 1132871701 Q * neofutur Ping timeout: 480 seconds 1132873188 P * vrwttnmtu Leaving 1132873225 J * neofutur ~neofutur@neofutur.net 1132874433 J * lilo_ ~lilo@lilo.usercloak.oftc.net 1132874512 Q * lilo Remote host closed the connection 1132875160 Q * sladen Ping timeout: 480 seconds 1132875405 J * sladen paul@starsky.19inch.net 1132875575 Q * Johnnie Ping timeout: 480 seconds 1132875920 J * NikDaPhreak ~NikDaPhre@193.24.241.34 1132876143 J * Johnnie ~john@acs-24-154-53-217.zoominternet.net