1195516800 Q * dowdle Remote host closed the connection 1195518508 M * bzed _Radiance: if ls freezes in /var/log, and in /proc.... I'm pretty sure you have a rootkit on your computer 1195518511 M * bzed not a good one, though 1195518553 M * _Radiance ok this is weird 1195518560 M * _Radiance i found something 1195518579 M * _Radiance the machine i'm connecting from to that server is the one causing the issue 1195518603 M * _Radiance when i ssh from that machine to ANY machine and i do ps aux for example then it freezes the connection ! 1195518609 M * _Radiance isn't this sick or what ? lol 1195518750 M * _Radiance i can do anything, but the moment i touch ps or try to ls /proc or ls /var/log/ it freezes the connection 1195518797 M * daniel_hozac try running yes... 1195518863 M * _Radiance ok lemme try 1195518944 M * _Radiance yes works hehe 1195518975 M * _Radiance so what could be causing this...never encountered this 1195519106 M * _Radiance doing a ps aux on the "evil" machine gives normal results 1195519121 M * _Radiance but ssh'ing to any other machine and doing it there causes the freeze 1195519193 M * _Radiance i'm going to restart that machine and see if it still occurs 1195519590 M * _Radiance hmm did not help 1195519595 M * _Radiance that machine is a vps btw 1195519608 M * daniel_hozac are you using vserver ... enter? 1195519680 M * _Radiance right now yes, same result in the vps 1195519692 M * daniel_hozac hmm? 1195519693 M * _Radiance but before i connected to that vps from a different machine using ssh 1195519738 M * _Radiance so what i did now is restarted the vps but still the same problem occurs 1195519758 M * _Radiance what i'll do now is restart the physical machine 1195519760 M * daniel_hozac i wouldn't really expect anything else... 1195519783 M * _Radiance since in this vps also nothing changed 1195519875 M * _Radiance but before i do i'll enter another vps on the same server and try 1195519877 M * _Radiance who knows 1195519934 M * _Radiance ok same result 1195520826 M * _Radiance even a reboot of the host didn't help 1195520832 M * _Radiance this is so weird 1195521038 Q * friendly12345 Quit: Leaving. 1195521076 J * friendly12345 ~friendly@ppp59-167-131-66.lns3.mel6.internode.on.net 1195522727 Q * FireEgl Quit: Bye... 1195523048 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 1195524575 Q * mire Ping timeout: 480 seconds 1195525388 Q * hparker Remote host closed the connection 1195525477 Q * cehteh Ping timeout: 480 seconds 1195526060 Q * AStorm Remote host closed the connection 1195526077 J * AStorm ~astralsto@tor-irc.dnsbl.oftc.net 1195526734 Q * FireEgl Quit: Bye... 1195530070 Q * balbir Ping timeout: 480 seconds 1195531972 Q * rorem- Ping timeout: 480 seconds 1195532028 J * mjeanson ~mjeanson@bas2-sherbrooke40-1128575760.dsl.bell.ca 1195532149 J * rorem- ~roremtank@bzq-219-46-202.isdn.bezeqint.net 1195533590 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 1195534090 J * Aiken ~james@ppp59-167-115-173.lns3.bne4.internode.on.net 1195534169 N * AStorm Guest1299 1195534175 J * AStorm ~astralsto@tor-irc.dnsbl.oftc.net 1195534214 Q * Guest1299 Remote host closed the connection 1195534537 Q * bragon Ping timeout: 480 seconds 1195534604 Q * Aiken Quit: Leaving 1195535851 Q * mjeanson Remote host closed the connection 1195537442 J * shuri ~shuri@64.235.209.226 1195538914 Q * duckx Ping timeout: 480 seconds 1195540139 J * duckx ~Duck@81.57.39.234 1195540462 Q * shuri Quit: Leaving 1195540644 J * virtuoso_ ~s0t0na@ppp91-122-94-135.pppoe.avangard-dsl.ru 1195540816 J * sharkjaw ~gab@shell.ormset.no 1195541054 Q * virtuoso Ping timeout: 480 seconds 1195541299 Q * sharkjaw Ping timeout: 480 seconds 1195541431 J * Aiken ~james@ppp59-167-115-173.lns3.bne4.internode.on.net 1195541902 J * JonB ~NoSuchUse@kg0-199.kollegiegaarden.dk 1195542070 J * sharkjaw ~gab@shell.ormset.no 1195542154 J * DavidS ~david@p54811FCE.dip0.t-ipconnect.de 1195542267 J * jazzanova ~boris@87.69.128.138 1195542269 M * jazzanova hi 1195542315 M * jazzanova i have a custom debian based fs. How can I put it in vserver ? 1195542526 M * JonB filesystem? 1195542626 M * jazzanova all files etc. 1195542641 M * jazzanova i have a debian-based usb stick 1195542642 M * JonB i do not understand 1195542652 M * jazzanova bootable usb stick 1195542659 M * jazzanova i want to turn it intot a vserver 1195542666 M * JonB guest or host? 1195542670 M * jazzanova guest 1195542694 M * JonB well, mount it, clean up /dev 1195542700 M * jazzanova my host already has vserver setup 1195542700 M * JonB make some configuration files for it 1195542710 M * JonB then start the guest 1195542738 J * Alikus ~alikus@217.150.200.212 1195542744 M * jazzanova and copy over ? 1195542772 M * JonB or run it from the USB stick 1195542797 M * jazzanova no, i don't want to run it from usb stick 1195542814 M * JonB then create a new guest 1195542815 M * jazzanova do i make a blank guest, and then delete /vservers/myguest and copy usb contents into it ? 1195542822 M * Alikus ! 1195542830 M * JonB keep the newguest /dev 1195542841 M * JonB and possibly /sys /proc ... 1195542846 M * JonB then copy the USB stick over 1195542865 M * jazzanova ok 1195542920 M * jazzanova how can I make a very simple guest ? 1195542935 M * JonB using the vserver newgust build command 1195542935 M * jazzanova a blank guest ? 1195543300 M * jazzanova like this: 1195543303 M * jazzanova vserver galt build --hostname galt --interface dummy0=eth0:10.0.0.124/8 --rootdir /var/lib/vservers 1195543310 M * jazzanova ? 1195543320 M * jazzanova galt is the new name 1195543514 M * jazzanova it says "no build method" specified. 1195543554 M * JonB read the documentation on the homepage 1195543575 M * jazzanova it wants -m option 1195543579 M * jazzanova but what do i give it 1195543585 M * jazzanova i don't want to download anything 1195543764 Q * JonB Quit: This computer has gone to sleep 1195543786 J * JonB ~NoSuchUse@kg0-199.kollegiegaarden.dk 1195543933 Q * JonB 1195545995 J * mire ~mire@245-169-222-85.adsl.verat.net 1195546034 Q * Aiken Quit: Leaving 1195546677 M * jazzanova hello, anyone alive ? how can I create a new guest without downloading anything ? i have the fs in a folder 1195546929 M * arachnist jazzanova: template, rsync or clone methods on vserver foobar build 1195546962 M * arachnist err, not template. template assumes taht you have the thing in a .tar.bz2 1195546988 M * harrydg [harry@lois ~]$ head -n5 ChangeLog-2.6.23.[2345678]|grep "Date: " 1195546988 M * harrydg Date: Fri Nov 16 08:19:12 2007 -0800 1195546988 M * harrydg Date: Fri Nov 16 08:24:58 2007 -0800 1195546988 M * harrydg Date: Fri Nov 16 08:36:42 2007 -0800 1195546988 M * harrydg Date: Fri Nov 16 09:27:48 2007 -0800 1195546990 M * harrydg Date: Fri Nov 16 09:33:58 2007 -0800 1195546993 M * harrydg Date: Fri Nov 16 09:43:31 2007 -0800 1195546995 M * harrydg Date: Fri Nov 16 10:14:27 2007 -0800 1195546998 M * harrydg moehaha... silly developers :) 1195547029 M * arachnist hmm? 1195547067 M * arachnist gazillion releases a day? 1195547068 M * arachnist :> 1195547093 J * larsivi ~larsivi@85.221.53.194 1195547205 J * dna ~dna@p54BCD309.dip.t-dialin.net 1195547802 J * gebura ~gebura@77.192.186.197 1195547991 M * sannes Think the stable team divided it up into such chunks to make it easier to pinpoint what would go wrong if anything went wrong :P 1195548558 M * gebura hi 1195549250 J * JonB ~NoSuchUse@kg0-199.kollegiegaarden.dk 1195549607 M * jazzanova arachnist: i can't find the documentation on the methods. 1195549631 M * jazzanova can you give me an example ? 1195549666 M * jazzanova i have all my files in a directory 1195549740 M * Loki|muh vserver build build --help 1195549745 M * Loki|muh this gives you a help 1195549758 M * Loki|muh maybe you want the skeleton method 1195549780 M * Loki|muh and the copy the files in the directory build with the skeleton method 1195549792 M * Loki|muh then* 1195550097 Q * FireEgl Read error: Connection reset by peer 1195550119 M * jazzanova whats a good way to copy files to presesrve stuff ? 1195550172 M * DavidS jazzanova: cp -a ? 1195550237 M * jazzanova ok 1195550241 M * jazzanova trying it 1195550678 M * jazzanova ok, i want to try to install debian vserver 1195550738 M * jazzanova there's not example for debian install, only ubuntu. 1195550791 M * Loki|muh the main difference between debian and ubuntu are different mirrors and release names 1195550876 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 1195551025 Q * JonB Ping timeout: 480 seconds 1195551220 M * jazzanova ERROR: Could not download the debootstrap package from 1195551221 M * jazzanova http://ftp.debian.org/debian/pool/main/d/debootstrap/debootstrap_1.0.3_all.deb 1195551514 M * jazzanova is this correct: 1195551520 M * jazzanova vserver rearden build --hostname rearden.hypervolume.com --interface eth0:10.0.0.125/24 --rootdir /var/lib/vservers -m debootstrap -- -d etch 1195551725 M * sid3windr what host system are you installing it on? 1195551729 M * jazzanova ubuntu 1195551736 M * sid3windr then just apt-get install debootstrap first 1195551761 M * jazzanova its a very old ubuntu system 1195551770 M * jazzanova repositories are gone 1195551779 M * jazzanova its hoary 1195551788 M * jazzanova can i download it manually ? 1195551804 M * Loki|muh yes 1195551816 M * jazzanova where ? 1195551823 M * Loki|muh packages.debian.org 1195551845 M * jazzanova last week i install a fresh vserver ubuntu, and it automatically downoladed deboostrap for me 1195551850 M * jazzanova how come it doesn't work this time. 1195551915 M * jazzanova where would I put the deboonstrap packgae ? do I download one from debian, and install it on the ubuntu:hoary system ? 1195551934 M * Loki|muh yes, dpkg -i file.deb 1195551972 M * jazzanova ok, its working 1195551979 M * jazzanova something is happening 1195552251 M * jazzanova 'vserver ... suexec' is supported for running vservers only; aborting... 1195552263 M * jazzanova after base system was intstalled, it crashed 1195552265 J * meandtheshell ~sa@85.127.105.52 1195552294 M * jazzanova vcontext: vc_ctx_create(): File exists 1195552296 M * jazzanova that too 1195552368 J * JonB ~NoSuchUse@130.227.63.19 1195552555 M * jazzanova boris@archimedes:~$ sudo vserver rearden start 1195552555 M * jazzanova WARNING: --nid is not supported by this version 1195552555 M * jazzanova vcontext: vc_ctx_create(): File exists 1195552586 M * jazzanova the --nid warning is not fatal, cause it happens for me on another working vserver. 1195552603 M * DavidS do you have unique context values? 1195552613 M * Loki|muh jazzanova: http://www.christianschenk.org/blog/upgrading-vserver-from-20-to-22/ 1195552642 M * jazzanova i have a recent vserver-utils 1195552662 M * jazzanova vserver 0.30.214 1195552675 M * jazzanova davids: what are context values ? 1195552712 M * DavidS /etc/vservers/*/context 1195552970 M * jazzanova ok, I have one existing working vserver with context 40002, and te new one got the same number. 1195552977 M * jazzanova should I make teh new one 40003 ? 1195553043 M * jazzanova i did it 1195553051 M * jazzanova looks like it's working. 1195553056 M * jazzanova thanks 1195553057 M * jazzanova :) 1195553481 M * JonB how well does vserver handle atime updates? 1195554216 J * cehteh ~ct@pipapo.org 1195555243 P * friendly12345 1195557197 N * _Radiance Radiance 1195558061 M * jazzanova i'm trying to get portforwad going into my guest 1195558065 M * jazzanova http://archimedes.hypervolume.com/~boris/t.txt 1195558071 M * jazzanova why doesn't it work 1195558092 M * jazzanova i'm forward 7061 on host to 7051 on guest (10.0.0.125) 1195558373 M * zbyniu telnet archimedes.hypervolume.com 7061 1195558373 M * zbyniu Trying 69.72.214.226.7061... 1195558373 M * zbyniu Connected to 69.72.214.226. 1195558373 M * zbyniu Escape character is '^]'. 1195558378 M * zbyniu it works 1195558547 M * jazzanova ok 1195558554 J * mark ~mark@ip-83-238-175-215.netia.com.pl 1195558577 M * jazzanova weird :) 1195558645 M * mark hello everyone 1195558668 M * mark does anybody know how to set caps for bind so it doesn't complain it can't bind UDP socket? 1195558683 M * mark I've set echo "NET_BIND_SERVICE" >> bcaps 1195558699 M * mark yet bind still can't bind to UDP port 53 1195558905 M * mark I've got: 1195558907 M * mark redirs (87.204.147.163) / % cat /proc/18873/status | grep Cap 1195558908 M * mark CapInh: 0000000000000000 1195558908 M * mark CapPrm: 00000000344c04ff 1195558908 M * mark CapEff: 0000000000000000 1195558926 M * mark on BIND pid 1195559005 M * jazzanova zbuyniu: it doesn't work well, its slow. i have softawre X that works well if i portforward with ssh -L, but not directly over iptables. 1195559033 M * jazzanova zbuyniu: i notice the same if i sseh directely (iptable doesn 2223->22) 1195559043 M * jazzanova after awhile connection gets stuck 1195559329 J * Aiken ~james@ppp59-167-115-173.lns3.bne4.internode.on.net 1195559380 M * zbyniu jazzanova: hmm, weird 1195559455 Q * ^Toad_ Ping timeout: 480 seconds 1195559468 M * zbyniu mark: are you sure it is caps problem? 1195559515 M * mark no 1195559520 M * mark I am not totally sure 1195559522 M * zbyniu mark: default (344c04ff) bind is permited 1195559525 M * jazzanova is it easy to change fake ip to real ip ? how do I do it ? 1195559536 M * mark hmm 1195559540 M * mark hold on a sec 1195559569 M * mark Nov 20 11:16:41 redirs named[18873]: could not listen on UDP socket: permission denied 1195559583 M * mark so I figured it's a caps problem 1195559589 M * mark but it doesn't have to be 1195559590 M * harrydg seems so 1195559613 M * harrydg is there something listening on 53/udp on the host itself? 1195559615 M * mark what I don't get is why there's a difference in BIND's effective caps 1195559636 M * mark I configured BIND on the main host to listen on another IP 1195559641 J * ^Toad ~tl@tyler.cs.brown.edu 1195559643 M * mark hold on I'll check that again 1195559680 M * zbyniu hmm, but named drops caps AFAIR, so maybe there is problem? 1195559753 M * mark zbyniu: yes, that's probably it 1195559764 M * mark question is, where do I go from here to get BIND functioning 1195559790 M * mark harrydg: I killed named on main host and restarted bind on guest and got the same answer 1195559803 M * harrydg ah 1195559805 M * mark so no, there's not another thing listening on 53 1195559836 M * harrydg mark: you should be careful that you don't start it on the host ... if it listens on all interfaces, your guest won't be able to bind to that socket 1195559852 M * harrydg play with caps a little 1195559860 M * zbyniu mark: echo CAP_SYS_RESOURCE >> bcapabilities 1195559864 M * harrydg try sysadmin cap, and see if that works 1195559869 M * harrydg work your way through ;) 1195559870 N * Bertl_zZ Bertl 1195559874 M * Bertl morning folks! 1195559878 M * mark hello Bert 1195559878 M * zbyniu mark: try that way 1195559879 M * harrydg da mastah has woken up1 1195559889 M * mark zbyniu, harrydg - Thanks! 1195559891 J * Yvo yvonne@vpn230.rz.uni-mannheim.de 1195559912 M * Bertl wb Yvo! ^Toad! 1195559924 M * Yvo thx! 1195559937 M * Bertl mark: problem is with bind? 1195559955 M * mark Bert, thanks for unleashing your considerable powers :-) 1195559958 M * mark while helping me 1195559963 M * mark anyway, it's probably caps problem 1195559976 M * mark or BIND dropping too much and unable to bind 53/UDP 1195559989 M * Bertl what kernel/patches do you use? 1195559994 M * mark cat /proc/18873/status | grep Cap 1195559994 M * mark [12:41] CapInh: 0000000000000000 1195559994 M * mark [12:41] CapPrm: 00000000344c04ff 1195559994 M * mark [12:41] CapEff: 0000000000000000 1195559997 M * mark standard debian 1195560011 M * mark 2.6.18-5-vserver-686 1195560030 M * mark what bothers me is that CapEff is all zeros 1195560039 M * Bertl hmm, no idea what patch that contains .. could be a kernel issue as well, you have bind running on the host? 1195560052 M * mark bertl: yes I do 1195560059 M * harrydg 12:56 < mark> harrydg: I killed named on main host and restarted bind on guest and got the same answer 1195560062 M * mark I mean bind is running on the host 1195560083 M * mark as harry pointed out 53/UDP is not taken 1195560094 M * Bertl if you check with 'lsof -ni :53' on the host, what do you see? 1195560097 M * Bertl (please use paste.linux-vserver.org for everything longer than 3 lines) 1195560103 M * mark ok 1195560157 M * mark http://paste.linux-vserver.org/9582 1195560164 Q * michal Ping timeout: 480 seconds 1195560226 M * mark there should be no conflict as I have guest on 87.204.147.163 1195560273 M * zbyniu mark: try CAP_SYS_RESOURCE 1195560308 M * Bertl mark: okay, could you 'strace -fF -o bind.log' the bind startup inside the guest? 1195560325 M * mark ok 1195560358 M * Bertl if possible upload that too (maybe just the part at the end) 1195560370 M * mark ok 1195560415 M * harrydg named 19430 bind 20u IPv4 51205 UDP 87.204.147.162:domain 1195560421 M * mark zbyniu hit the pot :-) 1195560422 M * harrydg ==> seems like 53/udp is in use 1195560424 M * mark it works 1195560430 M * mark CAP_SYS_RESOURCE 1195560434 M * harrydg but i'll shut up, Bertl willb e able to help you more 1195560473 M * mark harry: that's correct, but that is for another IP (162 last octet), while I use guest on 163 last octet 1195560483 M * mark bert, thanks 1195560509 M * mark the question I now have is setting CAP_SYS_RESOURCE == security hole? 1195560539 M * zbyniu mark: bind also setting ulimit() and often chroot() 1195560557 M * mark ?? 1195560558 M * harrydg i didn't have to add caps for my bind install @ server @ hetzner 1195560558 J * michal ~michal@www.rsbac.org 1195560579 M * zbyniu that is the reason 1195560613 M * mark bertl: should I do that strace (i mean is it any interesting for you) or is it redundant now? 1195560614 M * zbyniu harrydg: maybe some comlipation options? 1195560626 M * mark or harry's BIND runs as root 1195560652 M * mark mine is: /usr/sbin/named -u bind 1195560680 M * harrydg named 2948 0.0 0.1 43592 3588 ? Ssl Oct29 0:04 /usr/sbin/named -u named 1195560699 M * harrydg its a standard Centos5 install 1195560731 M * mark harrydg: do you have more than one BIND installation on several guests? 1195560738 M * harrydg nope 1195560741 M * harrydg only 1 1195560746 M * mark do you have BIND on the main host? 1195560751 M * harrydg why would i need more binds... 1195560757 M * harrydg i only need 1 dude, i only need 1 ;) 1195560770 M * mark well I need many, believe it or not, I have complicated setup 1195560772 M * Bertl mark: please run the strace, I just would like to see what issues are there (without the cap) 1195560777 M * mark ok, bert 1195560815 M * Bertl note: we know that bind is broken by design, and all recent Linux-VServer versions have workarounds for this specific app 1195560851 M * mark aha, ok - just a sec, I need to reconfigure it all 1195560875 M * Bertl I have no idea which version is in the debian kernels, and usually I suggest to disable the capability stuff in bind completely (which is why bind is misbehaving) 1195561062 M * Bertl note: if it is an older Linux-VServer release (2.0.x) what I assume, then you have three options to fix it: 1195561078 M * Bertl a) get a newer kernel (vs2.2.x branch) 1195561097 M * Bertl b) recompile the guest bind with --disable-capabilities 1195561123 M * Bertl c) add the CAP_RESOURCE to the guest (which usually _is_ a problem) 1195561127 M * Bertl http://linux-vserver.org/Capabilities_and_Flags 1195561161 M * Bertl scroll down to SYS_RESOURCE, to see what it means 1195561342 M * mark Errors 1195561342 M * mark * No code specified 1195561348 M * mark I get this from paste... 1195561350 M * mark sorry.. 1195561367 M * Bertl maybe too long or so 1195561368 J * yarihm ~yarihm@vpn-global-081-dhcp.ethz.ch 1195561376 P * Yvo 1195561398 M * Bertl look for the error value 1195561409 M * mark I managed to cut it down to: http://paste.linux-vserver.org/9583 1195561450 M * Bertl yeah, excellent, looks like the kernel is too old 1195561470 Q * gebura Read error: No route to host 1195561475 M * mark Bertl, I owe you 1195561510 M * mark another question - do you happen to have that patch for kernel for setting user/group quota inside a guest? 1195561520 M * mark a few days ago someone mentioned that to me 1195561537 M * mark that nobody was willing to test it or smth, - well I'm willing to test it 1195561556 J * gebura ~gebura@77.192.186.197 1195561588 M * Bertl sounds good, but I doubt that you are ... no offence meant, and I would really love to be proven wrong 1195561626 M * mark proficient in kernel hacking? :-) 1195561637 M * Bertl mark: thing is, we removed the _shared_ user/group quota some time ago, because nobody was willing/interested to do extensive testing 1195561682 M * Bertl mark: nevertheless, the user/group quota works fine on _separate_ partitions, e.g. LVM volumes 1195561704 M * mark I know, but I'd like to avoid LVM if possible 1195561724 M * Bertl mark: now, if you still want to volunteer for testing the _shared_ part, it will consume quite some time on your side 1195561729 M * mark I mean disklimits work fine fine 1195561733 M * mark hm 1195561744 M * mark ok, let me think about this 1195561749 M * mark right now I have my hands full 1195561761 M * mark but in a few months, after I get done with some projects, why not 1195561784 M * Bertl first, you need to setup a quota test scenario (for normal user/group quota) which test the quota correctness for all cases (including corner cases) 1195561810 M * Bertl this can be done without any Linux-VServer patch/kernel 1195561860 M * Bertl once that works, it has to be extended to handle the guest 'tags' in the tests, i.e. the cases where quota is transferred have to be tested too 1195561935 M * Bertl so, if you are good with scripting (any scripting language) you can be done in a weak for each step 1195561949 M * mark well I'm pretty good with scripting (bash, Python) 1195561958 M * mark I'm not very good with kernel innards though 1195561973 M * Bertl don't worry, that's my part :) 1195561977 M * mark ok 1195561995 M * mark I'm saving this chat, will get back to you when I have enough time to do this effectively 1195562017 M * Bertl okay, so surprise me and prove me wrong .. i.e. come back and do the work :) 1195562017 M * mark thanks a million, Bert 1195562029 M * mark I will, but no sooner than January 1195562032 M * Bertl you're welcome! feel free to hang around ... if you like 1195562035 M * mark diploma on my hands :-( 1195562060 M * mark this irc has been very helpful, thanks to all people here 1195562066 M * mark see you have to get back to work 1195562082 M * Bertl c-ya 1195562116 Q * mark Remote host closed the connection 1195563277 Q * Aiken Quit: Leaving 1195563614 Q * sharkjaw Ping timeout: 480 seconds 1195563732 Q * JonB Ping timeout: 480 seconds 1195564564 J * w1lm ~willem@a82-93-44-132.adsl.xs4all.nl 1195564744 J * Julius ~julius@p57B25102.dip.t-dialin.net 1195565309 J * julius_ ~julius@p57B25102.dip.t-dialin.net 1195565314 J * ema ~ema@rtfm.galliera.it 1195565559 Q * Julius Ping timeout: 480 seconds 1195565942 M * Bertl welcome w1lm! 1195565950 M * Bertl off for now ... back later ... 1195565955 N * Bertl Bertl_oO 1195565958 M * w1lm hey 1195566090 M * sid3windr :nr 64 1195566094 M * sid3windr (oops) 1195566675 J * lilalinux ~plasma@dslb-084-058-247-077.pools.arcor-ip.net 1195566839 J * mrfree ~mrfree@host1-89-static.40-88-b.business.telecomitalia.it 1195566878 M * mrfree hi all 1195566952 M * mrfree I'm reading vserver docs... but I haven't understand (at least) a thing... Can I set a public ip for a guest? Does the host acts like a bridge? 1195567014 M * sid3windr yes 1195567030 M * mrfree ok thz ;) 1195567268 Q * yarihm Ping timeout: 480 seconds 1195567306 J * yarihm ~yarihm@vpn-global-dhcp3-154.ethz.ch 1195567439 J * Punkie ~Punkie@goc.coolhousing.net 1195567468 J * JonB ~NoSuchUse@192.38.8.25 1195567518 M * w1lm Hey everybody 1195567559 M * w1lm I have read in the vserver wiki that a guest vserver can be a DHCP client.. but I haven't found any examples 1195567581 M * w1lm is that documented somewhere? 1195567888 J * shuri ~shuri@64.235.209.226 1195568002 Q * DavidS Quit: Leaving. 1195568118 J * DavidS ~david@p54811FCE.dip0.t-ipconnect.de 1195568186 M * mrfree but what if I want to remove a guest?? I can simple remove the guest dir into /vserver?? 1195568213 M * sid3windr yes 1195568214 M * sid3windr and the config 1195568225 M * mrfree ok... 1195568478 M * shuri hi there ltns! 1195568529 Q * yarihm Quit: Leaving 1195568659 J * moemoe moemoe@kuschelhoelle.netzhure.de 1195568687 M * moemoe hi guys, i have a little problem. i created a vserver with a private ip as described on http://linux-vserver.org/Networking_vserver_guests 1195568703 M * moemoe but the portforwarding doesn't work: https://nopaste.eof.name/296 1195568706 A * arekm want's to put existing debian machine as vserver guest. Does anyone know if debian init scripts need to be modified to work in vserver? 1195568712 M * moemoe any tips? 1195568746 M * moemoe arekm: i think there was a tool to adjust it, but can't remember the name 1195568808 M * moemoe arekm: stripserver inside vserver-debiantools 1195568837 M * arekm thanks, checking 1195568895 M * arekm that's it, thanks again 1195569086 M * arekm could not create shared memory segment: Invalid argument 1195569093 M * DavidS arekm: I have a puppet module targeted at debian in vserver, it also removes a few packages which are not neccessary 1195569123 M * arekm uh, what's the case with shm in guest? 1195569308 M * arekm http://pld.pastebin.com/f2f5d8f5c 1195569313 M * arekm shm* in guest differ 1195569346 J * DoberMann ~james@cap31-6-88-180-72-76.fbx.proxad.net 1195569354 Q * DoberMann 1195569372 M * arekm oh, it's in faq. shame on me 1195569440 M * arekm works :> 1195569481 J * DoberMann ~james@cap31-6-88-180-72-76.fbx.proxad.net 1195569582 Q * larsivi Quit: Konversation terminated! 1195570178 P * DoberMann 1195570579 Q * DavidS Quit: Leaving. 1195570593 J * DavidS ~david@p54811FCE.dip0.t-ipconnect.de 1195571279 J * julius__ ~julius@p57B272F4.dip.t-dialin.net 1195571712 Q * julius_ Ping timeout: 480 seconds 1195572515 J * ard6 ~ard@2002:d9c4:2909:1::1 1195573505 M * mrfree I noticed that my hosts obtain wrong dns infos... how can I change them? 1195573518 M * daniel_hozac hmm? 1195573579 Q * ex Ping timeout: 480 seconds 1195573599 M * mrfree I followed gentoo howto http://www.gentoo.org/proj/en/vps/vserver-howto.xml 1195573610 M * mrfree --hostname gentoo \ 1195573610 M * mrfree --interface eth0:192.168.1.253/24 \ and so on 1195573633 M * mrfree my guest has it's own public ip 1195573654 M * mrfree wait... 1195573691 M * mrfree ok ok my bad. the resolv.conf isn't auto-generated :) 1195573757 M * sid3windr isn't it copied from the host? 1195573757 J * ex ex@valis.net.pl 1195573963 M * daniel_hozac no. 1195573975 M * daniel_hozac it's copied from /etc/vservers/.defaults/files/resolv.conf, if that exists. 1195573989 M * sid3windr hmm! 1195574003 M * sid3windr and if it doesn't? 1195574014 M * daniel_hozac then you won't have one :) 1195574018 M * sid3windr now you got me wondering 1195574039 M * sid3windr I can't remember having created a new one in the last few created vservers 1195574042 M * sid3windr and I don't have the default one 1195574043 M * sid3windr odd; 1195574047 M * sid3windr I must be getting old 1195574061 M * daniel_hozac (well, you can have it in the distribution directory as well, but that seems rather pointless) 1195574626 J * dowdle ~dowdle@scott.coe.montana.edu 1195575336 J * larsivi ~larsivi@101.84-48-201.nextgentel.com 1195576018 J * the_hydra ~the_hydra@125.164.96.50 1195576021 P * w1lm 1195576260 Q * gebura Quit: Quitte 1195576762 J * dsoul_ darksoul@vice.ii.uj.edu.pl 1195577054 Q * dsoul Ping timeout: 480 seconds 1195577407 J * _gh_ ~gerrit@c-67-169-199-103.hsd1.or.comcast.net 1195577816 Q * JonB Quit: This computer has gone to sleep 1195578254 Q * ard Ping timeout: 480 seconds 1195578328 J * NetAsh ~na@87.247.101.146 1195578422 M * NetAsh hello 1195578447 M * daniel_hozac hi 1195578470 M * NetAsh can anyone explane me "sched" 1195578555 M * NetAsh http://www.nongnu.org/util-vserver/doc/conf/configuration.html#sched 1195578558 M * daniel_hozac i assume you've already read linux-vserver.org/CPU_Scheduler? 1195578563 M * NetAsh yap 1195578584 M * NetAsh i tried to dig everything possible befor asking :D 1195578598 M * daniel_hozac so, what about it then? 1195578640 M * NetAsh there are "idle" mode 1195578648 M * NetAsh or smthg like this 1195578689 M * NetAsh there are two sets of 1195578708 M * NetAsh fillrate / intervall 1195578752 M * NetAsh by that criteria it is chousen whitch to use (aka is idle or not) 1195578761 M * NetAsh so far as I understand 1195578769 M * NetAsh it shuld be somethig like this 1195578818 M * daniel_hozac if idle-time is enabled and the CPU is idle, R2 and T2 will be used instead of R1 and T1. 1195578822 M * NetAsh if system has more resorses to spare than normaly guest with its hard limit can consume 1195578833 M * NetAsh yap 1195578844 M * NetAsh idle = ? 1195578855 M * NetAsh cpu usage < 10% or that ? 1195578875 Q * mire Ping timeout: 480 seconds 1195578892 M * daniel_hozac no, idle is when there are no processes that want to use the CPU. 1195578930 M * NetAsh hmz 1195579008 M * NetAsh ok, if I set something like 1/4 (default) and 3/4 (idle) 1195579070 M * NetAsh ./ 1195579080 M * NetAsh ill beter try it, insted of asking :D 1195579128 M * NetAsh i'am trying to run vserver on the host with xorg running 1195579139 M * NetAsh and have some problems with mouse responsiviness 1195579173 M * NetAsh than i try compile something under the guest 1195579192 M * NetAsh so i decided to try limiting guest :D 1195579194 M * NetAsh ./ 1195579237 M * daniel_hozac should be fine... 1195579314 Q * meandtheshell Quit: Leaving. 1195579594 M * NetAsh this does not work at all 1195579612 M * NetAsh debian etch uses 2.0.X vserver 1195579620 M * NetAsh is this fieture in this version? 1195579717 M * daniel_hozac no, idle time is not in 2.0. 1195579756 M * NetAsh crap :D 1195579792 M * NetAsh priority-bias stands for? 1195579805 M * daniel_hozac a guest-wide nice value, essentially. 1195579828 M * NetAsh is this present in 2.0.X? 1195579854 M * daniel_hozac yes. 1195579859 M * NetAsh ok trying :D 1195579870 M * NetAsh il try to set it to 10 1195579949 M * arachnist jazzanova: vserver foobar build --help 1195579987 M * daniel_hozac you probably want -10. 10 would give it higher priority. 1195580016 M * NetAsh aka its value is inverse to nice value? 1195580022 J * bragon ~bragon@2001:7a8:aa58::1 1195580033 M * daniel_hozac yes, it's a priority. 1195580040 M * NetAsh ok thanks 1195580364 Q * lilalinux Remote host closed the connection 1195580377 M * NetAsh Token: 53 1195580377 M * NetAsh FillRate: 4 1195580377 M * NetAsh Interval: 5 1195580377 M * NetAsh TokensMin: 50 1195580377 M * NetAsh TokensMax: 60 1195580378 M * NetAsh PrioBias: -10 1195580415 M * NetAsh not much of the diference 1195580456 M * NetAsh :( 1195580538 M * NetAsh any sugestions how to run host with responsiv gui, and have cpu hungry guests at the same time? 1195580566 M * the_hydra enable full preemption on the host? 1195580583 M * NetAsh aaaa 1195580588 M * the_hydra instead of voluntary preemption or no preempt at all 1195580589 M * NetAsh and this is? how? 1195580604 M * the_hydra first, check your current kernel config 1195580618 M * NetAsh stok debian vith vserver 1195580632 M * the_hydra stock debian..ok, do you see /proc/config.gz? 1195580647 M * the_hydra in host, i mean 1195580654 Q * mrfree Ping timeout: 480 seconds 1195580669 M * the_hydra I refer to host in this context, just for clarification 1195580670 M * NetAsh i see /boot/config foo bar 1195580723 M * NetAsh that is options name? 1195580796 M * NetAsh grep $OPTION config 1195580804 M * NetAsh $OPTION = ? :D 1195580818 Q * DavidS Quit: Leaving. 1195580861 M * NetAsh CONFIG_PREEMPT_NONE=y 1195580861 M * NetAsh # CONFIG_PREEMPT_VOLUNTARY is not set 1195580861 M * NetAsh # CONFIG_PREEMPT is not set 1195580861 M * NetAsh # CONFIG_PREEMPT_BKL is not set 1195580912 M * the_hydra sory was away 1195580927 M * the_hydra shessh, preempt_none 1195580929 M * the_hydra no wonder 1195580930 M * NetAsh I need kernel recompile, or can i set it via proc/sys somether 1195580935 M * the_hydra recompile pls 1195580943 M * the_hydra and make it config_preempt 1195580948 M * NetAsh f..ck :D 1195580949 M * the_hydra and config_preempt_bkl 1195580953 M * the_hydra yeah, f$$k 1195580982 M * the_hydra also, make it HZ=1000hz for better GUI smooth response 1195581006 M * the_hydra full recipe: 1000 HZ + full preempt + preempt BKL + no_hz 1195581014 M * the_hydra daniel_hozac: pls CMIIW 1195581060 M * NetAsh thanks 1195581114 M * the_hydra IIRC no preempt is suggested for better compute server, but I think in your case it's different 1195581134 M * the_hydra unluckily, preempt mode is not something changable during runtime 1195581217 M * the_hydra I forgot which kernel version adopt the CFS scheduler, all I can say is, whenever possible use the very latest kernel 1195581232 M * daniel_hozac 2.6.23. 1195581238 M * the_hydra daniel_hozac: thanks man 1195581241 M * daniel_hozac there is no vserver support there though. 1195581247 M * the_hydra daniel_hozac: pfff 1195581251 M * the_hydra daniel_hozac: ok 1195581297 M * the_hydra NetAsh: alternative, go to kernel.kolivas.org and see if Staircase Deadline could fit with vserver patch in 2.6.22 1195581322 M * daniel_hozac highly unlikely... 1195581326 M * NetAsh ./ 1195581332 M * the_hydra daniel_hozac: pfff (again) 1195581347 M * the_hydra NetAsh: ok that's the best I can suggest..good luck 1195581361 M * NetAsh thanks 1195581388 M * NetAsh I personaly prefier debian stable kernels, becouse these are maintained 1195581399 M * daniel_hozac really? 1195581413 M * daniel_hozac is that why it includes several known bugs? 1195581413 M * NetAsh atleast it seems so :D 1195581426 M * the_hydra hehhehehe :D 1195581505 M * NetAsh damn, then it was the last time I compiled kernel? hmz :D 1195581519 M * the_hydra :) 1195581590 M * NetAsh I still have hope not to trach host, and do all compilations under guests 1195581885 M * NetAsh a 1195581888 M * NetAsh by the way 1195581948 M * NetAsh flower page (or how its named), could have tags like [2.0.x up] 1195581958 M * NetAsh [2.2.x up] 1195581960 M * NetAsh and so on 1195581982 M * daniel_hozac the utils should not specify kernel versions, no. 1195582011 M * daniel_hozac i will add the [added in 0.30.x] though, it's on my TODO-list. 1195582018 M * NetAsh yap, but fieture is dependant on kernel 1195582030 M * NetAsh or I am wrong? 1195582044 M * daniel_hozac but it's not the utils job to document that. 1195582047 Q * bragon Ping timeout: 480 seconds 1195582063 M * daniel_hozac http://linux-vserver.org/Feature_Matrix 1195582248 J * bragon ~bragon@2001:7a8:aa58::1 1195582265 J * Yvo ~yvonne@91.64.217.106 1195582318 Q * shuri Quit: This computer has gone to sleep 1195582340 J * shuri ~shuri@64.235.209.226 1195582373 M * NetAsh ./ 1195582396 M * NetAsh while kernel compiles, new questions arised :D 1195582417 M * NetAsh is it posible to change schedule parameters on the running guest ? 1195582471 M * daniel_hozac yes. 1195582472 Q * shuri Read error: Connection reset by peer 1195582533 M * NetAsh is it something horibly dificult, or echo XXX > somether? 1195582549 M * daniel_hozac vsched --help 1195582601 M * NetAsh cool 1195582620 M * NetAsh somehow missed such util is available :D 1195583556 M * faheem___ Hi. Is there an option to not have /tmp in a vserver guest mounted on tmpfs? 1195583612 M * daniel_hozac /etc/vservers//fstab 1195583802 M * faheem___ daniel_hozac: So... 1) Could I just remove the /tmp entry, and reboot? 2) Is it necesary to have /tmp as a separate partition for some reason? 3) Doesn't look like there is a default fstab. Where does it get the tmpfs stuff from? 4) Could I add a default fstab, and will it be recognized? 1195583817 J * hardwire` ~bip@rdbck-5696.palmer.mtaonline.net 1195583825 Q * hardwire Ping timeout: 480 seconds 1195583911 M * daniel_hozac 1) yes, 2) tmpfs increase performance when you have multiple guests using /tmp a lot, 3) /usr/lib*/util-vserver/defaults/fstab, 4) /etc/vservers/.defaults/fstab 1195583964 M * faheem___ daniel_hozac: Thanks, that 1195583975 M * faheem___ daniel_hozac: Thanks, that's very complete. :-) 1195583983 M * NetAsh but you have to have some spare ram :D 1195583994 M * daniel_hozac it's backed by swap too. 1195584019 M * NetAsh I dont know is it tmpfs problem 1195584070 M * NetAsh but than tmpfs hosts lots of files I saw some performance drop 1195584091 M * NetAsh mabe it was just a coincidence 1195584115 M * NetAsh newerless I use it my self :D 1195584126 M * faheem___ daniel_hozac: Can you elaborate on (2)? Is this any different from a normal machine, ie. using tmpfs on a normal system would presumably improve perormance too. 1195584139 M * NetAsh yes 1195584189 M * daniel_hozac faheem___: yes, it would. 1195584216 M * daniel_hozac faheem___: but with guests you have 10+ normal systems, typically all of them using the same disk... 1195584268 M * faheem___ daniel_hozac: So it is purely a performance issue, nothing to do with vserver itself? 1195584278 M * daniel_hozac right 1195584297 M * faheem___ daniel_hozac: Thanks. 1195584347 J * pmenier ~pmenier@ACaen-152-1-14-52.w83-115.abo.wanadoo.fr 1195584545 M * faheem___ I'm having a really weird problem with firefox. I have been running firefox from inside a 32bit vserver. Now the firefox on the host somehow has become the 32 bit version as well. 1195584576 M * the_hydra ??? hmm ??? 1195584591 M * faheem___ This seems to bizarre that I've triple checked this, and it seems to be the case. Basically, I can't use the firefox on my host system at all. 1195584591 Q * Punkie Quit: Odcházím 1195584592 M * the_hydra how do you know anyway? 1195584629 M * faheem___ the_hydra: a) It has the name of the guest on top when it starts up b) When I pull up the list of files to browse, they are the ones in the guest. 1195584640 M * faheem___ Specifically, in my home directory in the guest. 1195584661 M * faheem___ Maybe I'm doing something wacky. 1195584693 M * the_hydra sounds like somehow it shares the /home directory 1195584714 M * faheem___ I mean, I can type firefox on the console in the host, and it brings up the version from the 32 bit vserver guest. 1195584722 M * faheem___ the_hydra: No, completely separate systems. 1195584757 M * the_hydra wrong soft linking, maybe? 1195584759 J * hardwire ~bip@rdbck-2609.palmer.mtaonline.net 1195584770 Q * hardwire` Ping timeout: 480 seconds 1195584840 M * faheem___ the_hydra: I opened up a new terminal, and get my regular ff there, but on the already existing terminals, I get the 32 bit version. Weird. 1195584869 M * daniel_hozac echo $HOME 1195584878 M * daniel_hozac (and similar) 1195584915 M * the_hydra try to follow daniel_hozac's advice 1st 1195585021 J * ard ~ard@gw-tweakb16.kwaak.net 1195585032 J * JonB ~NoSuchUse@kg0-199.kollegiegaarden.dk 1195585101 M * faheem___ Ok, it stopped doing that once I closed the 32 bit firefox. Maybe it was trying to reuse a running firefox instance in some way. The 32 version came up pretty fast. 1195585110 M * faheem___ Env variables don't seem to be the problem... 1195585184 Q * ema Quit: leaving 1195585236 M * the_hydra hm ok 1195585578 J * hardwire` ~bip@rdbck-6096.palmer.mtaonline.net 1195585600 Q * hardwire Ping timeout: 480 seconds 1195586157 Q * pmenier Quit: Konversation terminated! 1195586215 J * shuri ~shuri@64.235.209.226 1195586429 M * NetAsh by 1195586432 Q * NetAsh Quit: Leaving 1195587487 M * shuri hi 1195587495 M * daniel_hozac hello 1195587522 M * shuri i try to move a old vserver 1195587526 M * shuri to another server 1195587537 M * shuri i got user/group permission problem on the new server 1195587558 M * daniel_hozac --numeric-ids 1195587565 M * daniel_hozac (if using rsync) 1195587617 M * faheem___ Well, this is a little OT. But my way of confirming my guess, http://weblogs.java.net/blog/felipeal/archive/linux/index.html. 1195587618 M * shuri allready got this 1195587620 M * shuri --numeric-ids 1195587626 M * faheem___ This seems a little wonky. 1195587632 Q * AStorm Remote host closed the connection 1195587648 M * shuri --perms 1195587664 M * shuri --owner 1195587669 M * shuri --group 1195587701 M * shuri -D --links --hard-links --times --block-size=2048 --recursive 1195587791 M * shuri daniel_hozac i need --numeric-ids or not? 1195587861 M * faheem___ Hmm, documented in man firefox, FWIW. 1195587866 M * daniel_hozac you do. 1195587874 Q * FireEgl Read error: Connection reset by peer 1195587884 M * shuri humm 1195587892 M * shuri i dont understand then 1195587922 Q * julius__ Remote host closed the connection 1195587932 M * shuri about my others rsync options? 1195587982 M * shuri --numeric-ids --perms --owner --group -D --links --hard-links --times --block-size=2048 --recursive 1195588449 J * AStorm ~astralsto@tor-irc.dnsbl.oftc.net 1195588727 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 1195589540 Q * Alikus Remote host closed the connection 1195589765 P * Yvo 1195590522 Q * Bertl_oO Remote host closed the connection 1195592943 T * * http://linux-vserver.org/ | latest stable 2.2.0.5, 2.0.3-rc3, devel 2.3.0.29, stable+grsec 2.2.0.5 | util-vserver-0.30.214 | libvserver-1.0.2 & vserver-utils-1.0.3 | He who asks a question is a fool for a minute; he who doesn't ask is a fool for a lifetime -- share the gained knowledge on the Wiki, and we'll forget about the minute ;) 1195592943 T * harry - 1195592961 J * Bertl herbert@IRC.13thfloor.at 1195593001 M * Bertl evening folks! 1195593031 A * Supaplex studies http://linux-vserver.org/CPU_Scheduler carefully 1195593205 M * fb evening Bertl :) 1195593447 J * eth01 ~ub3r@cpc2-stkp3-0-0-cust773.manc.cable.ntl.com 1195593451 P * eth01 1195593610 M * baldy_ hi all 1195593669 N * baldy_ baldy 1195594125 Q * sid3windr Server closed connection 1195594127 J * sid3windr luser@bastard-operator.from-hell.be 1195594146 M * baldy any1 know howtoi can limit the VSZ? (dont know what it is) 1195594204 J * mire ~mire@51-171-222-85.adsl.verat.net 1195594349 J * bzed_ ~bzed@devel.recluse.de 1195594385 Q * bzed Remote host closed the connection 1195594401 M * Bertl baldy: hey, you can limit the AS/VM 1195594443 M * baldy i setup a rss limit 1195594458 M * baldy can u tell me how can u limit that VSZ 1195594466 M * baldy and explain me what it exactly is? ;) 1195594467 M * Bertl the problem is this: 1195594486 M * Bertl normally you have linux running with overcommitment enabled 1195594509 M * Bertl this means, that processes can request whatever virtual memory they like 1195594522 M * Bertl regardless if that memory is actually available or not 1195594542 M * baldy the problem is 1195594546 M * Bertl then, when a process actually 'consumes' the memory 1195594567 M * Bertl and the memory is not available, you get a visit from the OOM killer 1195594580 M * baldy some customers need too much ram 1195594591 M * Bertl RAM != VM 1195594593 M * baldy and then the load increas to 40 andmore 1195594617 M * Bertl most likely not because of the RAM, more because of the fact that the ram isn't available :) 1195594618 M * baldy rss = ram? 1195594635 M * Supaplex so apparently I echo sched_hard R T > cflags for http://linux-vserver.org/CPU_Scheduler#Hard_Limit ? 1195594636 M * Bertl RSS is pages mapped in memory 1195594657 M * Bertl Supaplex: R T ? 1195594659 M * daniel_hozac Supaplex: echo sched_hard > flags, yes. 1195594694 M * Supaplex tokens R every interval T 1195594710 M * Supaplex ah flags. but how do I set the limit to? 1195594721 M * Bertl you put them into the scheduler config 1195594746 M * daniel_hozac http://www.nongnu.org/util-vserver/doc/conf/configuration.html#sched 1195594979 N * bzed_ bzed 1195595193 N * dsoul_ dsoul 1195595835 M * Bertl baldy: to get back to your issues, it looks like your system starts trashing, this will cause the high load 1195595961 J * emtt1 ~eric@dynamic-acs-24-154-85-144.zoominternet.net 1195595978 M * Bertl welcome emtt1! 1195595989 M * emtt1 Bertl: hey! 1195595993 M * emtt1 quick Q 1195596020 M * emtt1 any plans to implement setting the MAC on a guest "interface" 1195596039 M * daniel_hozac given that guests don't have interface, how would you expect that to work? 1195596044 M * daniel_hozac +english 1195596044 M * Bertl as there are no 'guest interfaces' no :) 1195596085 M * Bertl emtt1: Linux-VServer uses IP (layer 3) isolation 1195596141 M * emtt1 yep.. go it... k, so guess this is something that depends on the *hopefully* future layer 2 namespaces in vanilla kernel 1195596161 M * emtt1 what's the status on that? 1195596176 M * Bertl mainline is not there, AFAIK 1195596204 M * Bertl but once it will be in mainline, you can choose it as (slower) alterntive to the layer3 isolation 1195596208 M * daniel_hozac last i read, they were still working on getting the namespaces working with IPv6. 1195596237 M * emtt1 k, at least there is forward motion :) 1195596240 M * emtt1 tnx guys 1195596291 M * Bertl always, you're welcome! 1195596621 J * Jusic jusic@dslb-088-076-033-240.pools.arcor-ip.net 1195596664 M * Supaplex seems to be working for now, thanks guys. kontact in this vserver guest via ssh -fX, isn't starving my desktop now. thanks =) 1195596675 M * Bertl you're welcome! 1195596681 M * Bertl welcome Jusic! 1195596689 M * Jusic hi :) 1195596978 M * Supaplex all the dots are there. they're not easy to connect :P 1195599144 Q * Jusic 1195599197 Q * Vudumen Server closed connection 1195599198 J * Vudumen ~vudumen@perverz.hu 1195599292 J * Bertl_ herbert@IRC.13thfloor.at 1195599302 Q * Bertl_ 1195599447 Q * eviljonn1 Server closed connection 1195599447 J * eviljonny ~eviljonny@loki.eviljonnys.com 1195600290 Q * dna Quit: Verlassend 1195600531 Q * glen_ Server closed connection 1195600550 J * glen_ ~glen@elves.delfi.ee 1195600604 Q * cohan_ Server closed connection 1195600605 J * cohan ~cohan@koniczek.de 1195600666 Q * Radiance Server closed connection 1195600688 J * Radiance 41763ecf63@193.16.154.187 1195601334 Q * blizz_ Server closed connection 1195601334 J * blizz ~stephan@evilhackerdu.de 1195601463 M * jazzanova hello, anyone here ? 1195601477 M * Bertl sure, what's up? 1195601479 M * jazzanova i have a virtual ip on my vserver 1195601484 M * jazzanova hey Bertl 1195601490 M * jazzanova and i use dnat you gave me 1195601497 M * Bertl a virtual ip, sounds interesting ... 1195601498 M * jazzanova the thing is that it workes really bad 1195601510 M * jazzanova private subnet ip 1195601527 M * jazzanova here's example: http://www.rearden.flowersinvancouver.com:8087/faq.htmlhttp://www.rearden.flowersinvancouver.com:8087/faq.html 1195601528 M * Bertl how does it 'work bad' for you? 1195601533 M * jazzanova http://www.rearden.flowersinvancouver.com:8087/faq.html 1195601541 M * jazzanova so rearden is the vserver 1195601560 M * jazzanova it can't make the connection half the time 1195601564 M * jazzanova i notice the same with ssh 1195601579 M * jazzanova connecting to vhost works bad if i connect directly from outside 1195601587 M * jazzanova ok if i connect from the host 1195601595 M * jazzanova here's my dnat 1195601603 M * Bertl besides for the site engine being broken, it seems to work fine here 1195601622 M * Bertl why is it port 8087? 1195601630 M * jazzanova http://archimedes.hypervolume.com/~boris/t.txt 1195601655 M * Bertl use iptables -t nat -vnL 1195601731 M * jazzanova http://archimedes.hypervolume.com/~boris/tt.txt 1195601733 M * Bertl why does archimedes.hypervolume.com resolve to 127.0.0.1? 1195601740 P * Supaplex 1195601766 M * jazzanova here do you see that ? 1195601778 M * Bertl it looks to me like you are testing from the host or so? 1195601798 M * jazzanova im testing from my laptop, far from the host 1195601821 M * Bertl so what about the archimedes.hypervolume.com resolving to 127.0.0.1 then? 1195601822 M * jazzanova the dumps i sent you are done from the host (i sshed in) 1195601851 M * jazzanova yeah, thatst not good 1195601852 M * Bertl it looks to me like you created some name resolver issue on the host 1195601881 M * Bertl i.e. archimedes.hypervolume.com sometimes resolving to 69.72.214.226 but sometimes also to 127.0.0.1 1195601892 M * Bertl which of course will mess up things pretty bad 1195601902 M * jazzanova ok, so i fixed that part 1195601919 M * jazzanova now it will always resolve to 69* 1195601933 M * Bertl you might also want to specify the interfaces (network) in the DNAT rules 1195601958 M * Bertl to avoid DNAT-ing local connections, which should use the 10.0.0.125 and the proper ports 1195601994 M * Bertl i.e. add an -i eth0 to those rules 1195602140 M * Bertl the SNAT rule probably wants an -d ! 10.0.0.0/8 to avoid SNAT-ing traffic between 10.x.x.x addresses 1195602326 M * jazzanova http://archimedes.hypervolume.com/~boris/ttt.txt 1195602386 M * Bertl okay, so what problems do you see now? 1195602394 M * jazzanova Bertl: its better after the 127.0.0.1 fix 1195602407 M * jazzanova thats a good 1195602415 M * jazzanova i gotta play with it a bit 1195602424 M * Bertl okay, let us know how it goes ... 1195602461 M * jazzanova thanks! 1195602466 M * Bertl you're welcome! 1195602468 M * jazzanova you really spotted that one 1195602501 M * Bertl networking is tricky .. you have to look closely :) 1195602982 M * jazzanova ok, my ssh session just got stuck 1195603016 M * Bertl from where, to where? 1195603140 Q * arekm Server closed connection 1195603145 J * arekm arekm@carme.pld-linux.org