1136766900 M * dothebart YES. gotcha. 1136767063 M * dothebart Processor type and features -> Firmware Drivers -> BIOS update support for DELL systems via sysfs 1136767086 M * dothebart this was the bad guy... tnx _are_ 1136767699 J * bbd evaldas@barclay.balt.net 1136768008 M * bbd i know this may sound stupid but what is the best to way to let vserver to notify host to run some command? host is not running any other daemon besides vserver 1136768099 M * bbd write some tiny daemon to listen on specified port? does netlink work between host/vserver? 1136768405 M * daniel_hozac bbd: UNIX socket would work too. 1136768416 M * daniel_hozac sounds _really_ scary though. 1136768438 M * daniel_hozac why do you need to run commands on the host from the vserver? 1136768490 M * bbd or to run on another vserver. here's my example. we have 2 vservers - apache and postfix. i need to calculate size of /home/vmail/domain on apache vserver which doesn't have mounted /home/vmail for security reasons 1136768513 J * Cru ~mindwarp@bastardrouterfromhell.e.de.wahlich.com 1136768514 M * Cru morning 1136768521 M * daniel_hozac morning. 1136768534 M * bbd socket is an idea too. 1136768537 M * daniel_hozac another vserver should be doable with just ssh. 1136768575 P * matto 1136768835 M * bbd i'm not running ssh on it. btw i have an idea to implement it faster using netcat.. 1136769104 M * daniel_hozac Cru: would adding gpgkey=http://naturidentisch.de/packages/cru.asc to the .repo be correct? (i.e. is that the key used?) 1136769132 M * Cru it is 1136769154 M * daniel_hozac Cru: might want to change the name too, [cru-vps] would more accurate. 1136769211 M * Cru well, the packages were created for CentOS 4, although they run on a number of other RH based distros as well 1136769276 M * Cru I could create a fc4-vps repo etc. 1136769286 M * daniel_hozac right, but [base] might override the actual base repository. 1136769294 M * daniel_hozac (in the .repo) 1136769361 M * Cru hmm, sorry, have no idea of .rpeo files, I am used to work with apt, yum support was added just because livna dropped apt support, so many people switched to yum 1136769507 Q * Andi Quit: Chatzilla 0.9.68.5.1 [Firefox 1.0.7/20050922] 1136769737 M * daniel_hozac would you mind if i included the .repo in my util-vserver packages? 1136769804 M * daniel_hozac (would save the typical install useless packages, replace with your RPMs procedure and just install your packages directly) 1136769812 M * Cru feel free to use it ;) 1136770064 M * daniel_hozac oh, and on /packages, it says Linux Virtual Servers, that's another project ;) 1136770305 M * Cru heh, okay, will try to mind to change that 1136771812 J * Smutje ~Smutje@xdsl-87-78-58-89.netcologne.de 1136771920 Q * Smutje_ Ping timeout: 480 seconds 1136774024 M * Bertl hmm .. cools 2 power outages and network troubles as well ... 1136774033 M * Bertl evening folks, btw! 1136774105 A * Bertl is reading up ... 1136774376 M * Bertl tso: well, they have not been updated and/or adapted to any changes since a long time, they create 'known' broken guests and they do provide functionality which is already there in the mainline tools, so what is the purpose except for adding unnecessary work? 1136774421 M * Bertl waldi: I've no idea regarding 0.30.204 on ppc64, but on ppc they work quite fine (of course, not the debian ones) 1136774577 M * Cru moin bertl 1136774801 M * Bertl hey Cru! everything fine? 1136774871 M * Cru sure - currently I wonder whether POSIX.pm exists on Perl for Windows... ;) 1136774889 M * Cru have to Wintendo box to check that... 1136774942 M * Cru for the meaning of POSIX it should not exist, but who knows... ;) 1136774994 M * daniel_hozac i thought Windows was mostly POSIX compliant. 1136775024 M * Bertl hmm, well, I'm sure if you read the details on the 'M$ Services for Unix' pages then you will find that POSIX is a M$ invention :) 1136775100 M * Cru heh, okay, I will assume Wintendo to support strftime ;) 1136775529 J * ^_^ ~aaaaaaaaa@218.111.212.240 1136775538 N * ^_^ Guest1478 1136775549 M * Cru POSIX Subsystem seems some Windows extra-foo that exists for NT/2K but not for current versions 1136775558 M * Guest1478 WHO WAN FUCK????? 1136775560 P * Guest1478 1136775610 M * daniel_hozac haha, typical. 1136775853 M * Cru they simply sell it separately now, making more $$$ is even more typical ;) 1136776766 M * Cru okay, gotta go to bed now 1136776774 M * Cru bbl 1136776789 Q * Cru Quit: use Unices; $live->free() || die; 1136778579 Q * matti Quit: 8-X 1136779225 M * Bertl off to bed now ... c'ya tomorrow! 1136779242 N * Bertl Bertl_zZ 1136779306 J * balbir ~balbir@59.92.152.105 1136781080 Q * NikDaPhreak Quit: Hybernating my brain.... 1136781590 Q * flock Ping timeout: 480 seconds 1136781756 J * flock ~restless@l192-117-111-12.broadband.actcom.net.il 1136782906 P * undefined 1136785490 J * NikDaPhreak ~NikDaPhre@217.75.141.95 1136785494 M * NikDaPhreak hi all 1136786231 J * coocoon ~kvirc@p54A07C08.dip.t-dialin.net 1136788875 Q * Johnnie Read error: Connection reset by peer 1136789770 J * menomc ~amery@200.75.27.122 1136789880 Q * mnemoc Ping timeout: 480 seconds 1136789882 N * menomc mnemoc 1136790369 P * coocoon So Long, and Thanks for All the Fish! 1136792912 J * coocoon ~kvirc@p54A07C08.dip.t-dialin.net 1136792962 M * coocoon hello is someone there for helping me to access my vserver over ssh and knows a lot about iptables 1136793090 M * NikDaPhreak coocoon hi 1136793128 M * coocoon hi 1136793140 M * NikDaPhreak coocoon: i might try to help. what is the problem? 1136793156 M * coocoon we have talked last times haven't we 1136793180 M * NikDaPhreak coocoon: not sure 1136793212 M * coocoon ok I have solved a problem to access the network with a nat rule over ip 1136793225 M * coocoon but i can't access my vserver over ssh 1136793233 M * coocoon i have configured the port2222 1136793252 M * NikDaPhreak coocoon: well... you will need to DNAT the incomming connections to the vserver ip 1136793259 M * coocoon in the sshd.conf of the vserver have set the ip of the vserver in the sshd.conf of the vserver 1136793266 M * coocoon have set the host ip and port 22 1136793273 M * coocoon in may host sshd.conf 1136793338 M * coocoon i ve got a command 1136793352 M * coocoon to solve this after that it worked but no access to my vserver 1136793353 M * NikDaPhreak coocoon: iptables -t nat -A PREROUTING -p tcp --destination-port 2222 -j DNAT --to-destination :22 1136793368 M * coocoon oh aha looks great 1136793394 M * coocoon vserverIP:22 1136793405 M * coocoon when i have set 2222 in the sshd conf 1136793415 M * NikDaPhreak coocoon: no. 1136793419 M * coocoon of the vserver sshd.conf 1136793436 M * NikDaPhreak coocoon: this would redirect port 2222 on the host to port 22 on the guest 1136793468 M * coocoon ok u mean i should let ssh port and do not set it to 2222 1136793478 M * coocoon i do niot undersatnd 1136793482 M * coocoon understand 1136793522 M * coocoon i have configured host to 22 1136793527 M * coocoon and the vserver to 2222 1136793551 M * NikDaPhreak coocoon: ok, than you do not need the port to be redirected - skip :22 1136793560 M * coocoon ok 1136793591 M * coocoon but maybe it s better to set the host port to another port as 22 1136793595 M * coocoon i have read this 1136793662 M * NikDaPhreak coocoon: if only a handfull of people manage the host (which sould be the case) it is generally a good idea to use nonstandard ssh port and key authentification with no plain password logins for security reasons 1136793684 M * NikDaPhreak coocoon: it is pretty stupid to wake up with a hacked host ;-) 1136793716 M * coocoon yes thats right 1136793744 M * coocoon i have a question 1136793772 M * NikDaPhreak I have a bunch of questions ;-) 1136793780 M * NikDaPhreak just ask ;-) 1136793802 M * coocoon is it possible to access all services over one ip from the to all vservers with a local ip behind the firewall 1136793815 M * coocoon i have read that masquerading makes it impossible 1136793825 M * coocoon host 1136793845 M * NikDaPhreak coocoon: if you want to use the same ports for more than one vserver - no, it is not possible 1136793874 M * NikDaPhreak coocoon: simple reason - each daemon/service needs to lock the port on the IP for itself 1136793877 M * coocoon no the ports 1136793884 M * coocoon can vary 1136793903 M * coocoon but i have got an infoermation to use a tool like this 1136793906 M * coocoon SysCP 1136793920 M * coocoon to accessing dns and httpd and more 1136793927 M * NikDaPhreak coocoon: ok, than it is perfectly possible 1136793949 M * coocoon do u use that 1136793952 M * coocoon tool 1136793961 M * NikDaPhreak coocoon: if you setup 3 vservers and use http on 81, 82 and 83 respectively - no problems 1136793972 M * coocoon oh ok 1136793980 M * coocoon thats my solution 1136793982 M * NikDaPhreak coocoon: but if you try to use http on port 80 on all vservers - no, it will not work 1136793983 M * coocoon ;-) 1136793991 M * coocoon but must i use this 1136793996 M * coocoon in that way 1136794014 M * coocoon i don 't think so 1136794046 M * NikDaPhreak coocoon: you mean you have to access the services on all the 3 vservers on the same port? no NAT rule can do that for you. 1136794067 M * coocoon no is it a bad wy 1136794074 M * coocoon way to use other prots 1136794079 M * coocoon i mean for the folks 1136794113 M * Hollow NikDaPhreak: you have to specify the netmask as /255.255.255.0 not /24 1136794119 M * Hollow haven't implemented CIDR yet 1136794135 M * NikDaPhreak coocoon: imagine you have 3 rooms (e.g. httpd service on 3 vservers). you need 3 doors (ports on the host/IP) - one for each room. 1136794169 M * NikDaPhreak Hollow: I'll test it in a second :-) 1136794185 M * coocoon ok thanx i have understand 1136794187 M * Hollow great.. you see, the network stuff is still a bit rough ;) 1136794241 M * coocoon another questions if i set v_sshd i got the message that chbind can't be found 1136794248 M * coocoon using bind9 1136794258 M * NikDaPhreak Hollow: vncontext -A -n 10002 -a 10.0.1.2/255.255.255.0 1136794261 M * coocoon but after reinstalling bind the same problem 1136794268 M * NikDaPhreak Hollow:cat /proc/virtnet/10002/info 1136794268 M * NikDaPhreak SiD3WiNDR: 10002 1136794268 M * NikDaPhreak Info: cda14ae0 1136794268 M * NikDaPhreak 0: 10.0.1.2/0.0.0.2 1136794281 M * Hollow hm 1136794290 A * Hollow goes looking 1136794311 J * schellh ~bla@ipsio492.ipsi.fraunhofer.de 1136794313 M * schellh hi 1136794330 M * NikDaPhreak Hollow: that's what I mean. U use a func converting the netmask. I guess it's behaving funny with your input... 1136794396 M * NikDaPhreak coocoon: enable Legacy in your kernel or use newer tools (which I am currently trying to figure out how to use) 1136794398 M * coocoon thanx phreak for the informations 1136794418 M * coocoon ok i will do this 1136794425 M * coocoon how long will u be here 1136794431 M * Hollow NikDaPhreak: have to compile the utils first ;) 1136794434 M * coocoon if i have questions about iptables 1136794438 M * NikDaPhreak coocoon: and you have to recompile bind or use some flags to allow it to bind to the raw interface 1136794452 M * coocoon ok i will do this 1136794456 M * schellh i have a sparc with kernel 2.6.14.4 and vserver 2.1 - when i start a freshly bootstrapped debian vserver, i get: wait(): No child processes 1136794457 M * NikDaPhreak Hollow: don't tell me you do not use your won utils ;-) 1136794471 M * Hollow NikDaPhreak: i don't have installed any utils on my dev box atm ;) 1136794500 M * coocoon but which newer tools 1136794514 M * coocoon u mean util-vserver tool or bind tools 1136794523 M * coocoon i am using sarge 3.1 stable 1136794534 M * NikDaPhreak coocoon: I do not know... did not get much sleep lately.. 1136794543 M * coocoon oh aha 1136794550 M * coocoon u r in germany 1136794553 M * coocoon or where 1136794576 M * NikDaPhreak coocoon a bit further more east ;-) 1136794613 M * coocoon oh aha 1136794669 M * coocoon na supi 1136794678 M * coocoon but i think it is better to speak in english 1136794681 M * coocoon ;-) 1136794694 M * coocoon maybe i have some questions for other users 1136794701 M * NikDaPhreak coocoon: private messages and channel are 2 different things ;-) 1136794705 M * coocoon who can't speak german 1136794727 M * coocoon oh ok 1136794734 M * NikDaPhreak coocoon: ;-) 1136794758 M * coocoon but which tools could be too old 1136794777 M * coocoon he chbind 1136794783 M * coocoon for the 1136794848 M * NikDaPhreak coocoon: chbind is the older method to virtualize the network interfaces. Yesterday evening I started to study Hollow's vserver-utils-1.0.1 to find a bug ;-) 1136794852 M * Hollow NikDaPhreak: hm, first.. it uses CIDR woho! 1136794862 M * Hollow second... the mask calculation is wrong 1136794885 M * NikDaPhreak coocoon: I still do not know how to isolate an IP address for a context 1136794895 M * NikDaPhreak Hollow: well... ;-) 1136794938 M * Hollow NikDaPhreak: beside the network thing... do you think it's usable? 1136794975 Q * shedi Quit: Leaving 1136795026 M * coocoon oh aha so I will try it also with this new tool 1136795089 M * coocoon or is it beta 1136795093 M * NikDaPhreak Hollow: well.. it's more virtuozzo syntax, but it's usable :-) 1136795095 M * Hollow yup 1136795120 M * coocoon where to find it 1136795129 M * NikDaPhreak coocoon: well most nice things more or less beta ;-) 1136795149 M * coocoon which kernel do u recommend 1136795155 M * Hollow NikDaPhreak: hm, well i changed the order on the command line because it's easier to implement... e.g. the prcfs command does not need the name argument, and i think there can probably be more commands like this where you have total different syntax 1136795205 M * NikDaPhreak Hollow: BTW is/should it be possible to "chbind" with vnconfig? I spend a few hrs yesterday with no success on that.. 1136795230 M * NikDaPhreak Hollow: I do not mind the syntax :-) I have to use virtuosso as well :-) 1136795251 M * NikDaPhreak s /virtuosso/virtuozzo/ 1136795274 M * Hollow NikDaPhreak: yup, sth like vncontext -C -n 123 -- vncontext -A -n 123 -a 192.168.0.1/24 -- httpd should do it 1136795277 M * Hollow (in theory ;) 1136795329 M * NikDaPhreak Hollow: vcontext...... -- vserver start test wold then "chbind" the whole vserver, or not? 1136795338 M * NikDaPhreak Hollow: again in theory ;-) 1136795372 M * Hollow yup, just like chbind does on vserver foo enter with the old utils 1136795386 Q * sladen Ping timeout: 480 seconds 1136795427 M * NikDaPhreak Hollow: well... it did not work yesterday... 1136795455 M * NikDaPhreak Hollow: it might be due to the netmask problem.. 1136795464 Q * _are_ Quit: bbl 1136795468 M * Hollow i hate these mask and bitwise things.. bah ;) 1136795483 M * NikDaPhreak Hollow: :-) 1136795515 M * NikDaPhreak Hollow: I am kind of developing a sixt sence in routing/networking ;-) 1136795539 M * NikDaPhreak Hollow: but... I'm not a programmer... 1136795578 M * schellh i have a sparc with gentoo ,kernel 2.6.14.4 and vserver 2.1 - when i start a freshly installed via bootstrap debian vserver, i get: wait(): No child processes 1136795590 M * schellh can someone help me with that ? 1136795674 M * schellh a strace often tells: = -1 ESRCH (No such process) 1136795741 M * schellh i am missing the contents of /vserver/name/etc/... 1136795818 M * NikDaPhreak Hollow: jist an observation I made - if I use /255 or /255.255.255.0 - it makes no differense. You somehow manage to get rid of everyting behind the first dot and the dot itself... 1136796078 J * eugenesan eugenesan@bzq-219-189-188.cablep.bezeqint.net 1136796137 M * eugenesan hi all, what is the latest kernel available for patching by 2.0.1? 1136796438 J * sladen paul@starsky.19inch.net 1136796680 J * NetAsh ~NetAsh@193.219.160.108 1136796866 M * schellh 2.6.14.3 ?! 1136796896 J * Milf ~Miranda@ipsio16.ipsi.fraunhofer.de 1136797203 M * harry 2.6.14.4 works as well 1136797220 M * harry the only error is the makefile VERSION= 1136797491 M * NetAsh hello everyone 1136797530 M * NetAsh somehow I did not expected such a big company in there 1136797592 M * NetAsh the real question is - how many are not sleeping at the moment ;) 1136797603 M * NetAsh ./ 1136797611 M * harry me! 1136797617 M * NetAsh cool 1136797638 M * schellh haha 1136797668 M * NetAsh I have some fouts regarding min/soft/hard limits 1136797694 M * NetAsh (btv my english gramar is not perfect) 1136797721 A * harry will help a collegue now 1136797727 M * harry cyall later 1136797740 M * schellh i am not familiar with this :( 1136797752 M * NetAsh its not the problem 1136797808 M * NetAsh it would be the best to diskus this with someone who myght implement 1136797823 M * schellh i agree hehe 1136797864 M * NetAsh ./ 1136797912 M * NetAsh are there the most from the amrecian continent, or europe or else ? 1136797954 M * NikDaPhreak NetAsh: mostly Germany/Europe 1136797976 M * NikDaPhreak NetAsh: what's bothering you about limits? 1136798174 M * NetAsh I read vservers mailing lists, googled a lot 1136798200 M * NetAsh and practikaly found no solution for confortable way to limit memory 1136798249 M * NetAsh I do not know is it esealy implementable (sorry, regardles I am developer - my c and c++ noleje vaided long ago ;( 1136798264 M * NetAsh so 1136798286 M * NetAsh supouse it is inmplemented 1136798342 M * NetAsh we have a system with 1 gig of ram and 2 gigs of swap 1136798347 M * NikDaPhreak NetAsh: well, you only can set hard limits. soft limits are not implemented yet 1136798355 M * NetAsh I know 1136798399 M * NetAsh this is why I propose a mechanism min soft hard shuld work on memory limits 1136798506 M * NetAsh is anyone interested 1136798523 J * shedi ~siggi@tolvudeild-205.lhi.is 1136798539 M * schellh surely it would be good, but i guess its hard to implement because of processes growing in memory 1136798561 M * NetAsh ok 1136798563 M * schellh there must be a kernel level daemon watching the usage, and the daemon would have to kill processes ? 1136798579 M * NetAsh scenario 1 1136798579 M * schellh i only suppose this, i am not a vserver dev 1136798616 M * NetAsh rss now acounts for guest using real ram 1136798685 M * NetAsh lets say we add new tipe of ressourse vrss - guest using real ram & swap 1136798742 M * NetAsh if we still have only hard limits 1136798785 M * NetAsh this would be good 1136798821 M * schellh well howto handle growing processes then ? some out of memory emergency exit or just kill them ? 1136798840 M * schellh i think for such hard limits its better using UML or something like it 1136798969 J * id23 ~id@p54A04588.dip0.t-ipconnect.de 1136798972 M * id23 Hi 1136798990 M * schellh hi 1136799126 M * NetAsh schellh imagine you have system with 1gb of ram and 1gb swap 1136799212 M * NetAsh you have 5 guests with 512MB vrss (size in swap & ram) hard limit 1136799223 M * schellh i know it can get very slow! but i also think on howto implement this limits. best would be some kernel level thing like a daemon accounts ram usage of the different contexts and denies further ram usage so the vservers process which wants more gets some out of memory error 1136799294 M * NetAsh arghz 1136799314 M * NetAsh I gues I slept too litle or I am too slow today :) 1136799333 M * schellh better ill shut up and you wait for some dev :) 1136799356 M * NetAsh not neceserely 1136799403 M * NetAsh in case you understand a concept - you myght forsee some bugy points in my sugestion 1136799461 M * NetAsh ./ 1136799478 M * NetAsh schellh "some dev" acounts for ?:) 1136799497 M * NetAsh aka what nicks are dev's 1136799575 J * _are_ ~are@62.112.159.81 1136799598 M * NikDaPhreak NetAsh: Bertl is the main maintainer and developer of the kernel patches and Enrico and Hollow - for the two branches of the utilities 1136799664 M * _are_ morning 1136799678 M * schellh hi 1136799687 M * _are_ btw, is there a debian package for Hollows toolversion? 1136799958 M * NetAsh util-vserver vserver-debiantools kernel-patch-vserver and thats all (atleast in official mirrors) 1136800100 M * _are_ I know they are not in official mirros 1136800116 M * _are_ util-vserver there is the legacy-networking version 1136800125 Q * dothebart Read error: Connection reset by peer 1136800128 J * dothebart ~willi@xdsl-81-173-227-17.netcologne.de 1136800179 M * _are_ looked into it and tried patching myself, but got lost in the documentation/source for lacking some understanding of the vserver technique in the kernel code. And now it seems Holow did what I tried to. .-) 1136800251 M * NetAsh I am quite new to vserver, but so far this triplet does all I need 1136800276 M * _are_ well, did for me, but i tried and dd interface #17 to a vserver a few days ago :-> 1136800784 J * prae ~prae@ezoffice.mandriva.com 1136801087 M * NikDaPhreak Hollow: ping 1136801094 M * dothebart _are_: read? i solved the riddle. 1136801127 M * _are_ dothebart: no, have not read it, what had been the matter? 1136801145 M * dothebart Processor type and features -> Firmware Drivers -> BIOS update support for DELL systems via sysfs 1136801152 M * dothebart this was the bad guy. 1136801229 M * _are_ :-> 1136801281 M * dothebart took me a couple of hours to find it out at all... i first was thinking of something in userspace, or that there was a bad rootkit installed or so... 1136801368 A * NetAsh out 1136801370 M * NetAsh by 1136801374 Q * NetAsh Quit: 1136801456 Q * balbir Ping timeout: 480 seconds 1136803545 J * balbir ~balbir@59.92.159.79 1136803699 Q * prae Quit: Execute Order 69 ! 1136804289 J * Andi ~Andi@socks.tuwien.ac.at 1136804606 J * prae ~prae@ezoffice.mandriva.com 1136805051 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1136805083 J * coocoon ~kvirc@p54A07C08.dip.t-dialin.net 1136806073 M * Hollow NikDaPhreak: sorry, was back at school, will look now at the net thing 1136806082 M * NikDaPhreak Hollow: :-) 1136806091 M * NikDaPhreak Hollow: i guess I fixed it ;-) 1136806101 M * Hollow ah, very nice, do you have a patch? 1136806114 M * NikDaPhreak *mask = atoi(addr_mask); 1136806114 M * NikDaPhreak *mask = htonl(*mask); 1136806114 M * NikDaPhreak if (inet_aton(addr_mask, &ib) == -1) 1136806115 M * NikDaPhreak return -1; 1136806115 M * NikDaPhreak *mask = ib.s_addr; 1136806127 M * NikDaPhreak oops... the comments are gone 1136806136 M * NikDaPhreak first 2 lines were commented out 1136806170 M * NikDaPhreak Hollow: no warranty of any kind ;-) 1136806188 M * Hollow which lines? coul you pastebin it? 1136806191 M * NikDaPhreak Hollow: it prints the correct netmask in /proc... but if it works.... 1136806208 M * NikDaPhreak Hollow: I'll make a diff 1136806360 M * NikDaPhreak Hol: the diff is not that big.. see private message.. was quicker ;-) 1136806411 M * Hollow ah, ic 1136806413 M * Hollow nifty 1136806443 M * NikDaPhreak Hollow: as I mentioned - I'm not really a programmer ;-) 1136806515 M * NikDaPhreak Hollow: the sollution was some copy/paste ;-) 1136806578 M * Hollow ah well, this is not for CIDR, right? 1136806621 M * NikDaPhreak Hollow: CIDR? 1136806670 M * Hollow classless interdomain routing 1136806685 M * Hollow e.g. 1.2.3.4/24 1136806691 M * Hollow 24 is the cidr notation 1136806695 M * NikDaPhreak Hollow: you said one needs e.g. 10.0.0.1/255.255.255.0 as address... so I did it based on this asumption 1136806698 M * Hollow 255.255.255.0 is the standard netmask notation 1136806714 M * Hollow yeah, best would be to support both ways ;) 1136806733 M * NikDaPhreak Hollow: yes, I know the difference... had only 4 hrs sleep tonight and too much blud in the caffaine ;-) 1136806749 M * Hollow heh 1136807714 M * Hollow NikDaPhreak: ok, seems to work now 1136807724 M * NikDaPhreak Hollow: :-) 1136807759 M * Hollow NikDaPhreak: http://home.xnull.de/misc/vncontext.patch 1136807768 M * NikDaPhreak Hollow: I still have the problem, that the new context has access to all the IPs and interfaces... 1136807819 M * NikDaPhreak Hollow: nice :-) 1136807837 M * Hollow i should strip the printf's :P 1136807876 M * Hollow NikDaPhreak: how did you test it? 1136807975 M * NikDaPhreak Hollow: vncontext -C -n 101 -- vserver start test ; vncontext -A -n 101 -a 10.0.0.10/255.255.255.0 ; vserver enter test 1136807998 M * NikDaPhreak Hollow: ifconfig shows all interfaces 1136808029 M * Hollow yeah, expected behaviour ;) 1136808058 M * Hollow you have to use vcontext too, with HIDE_NETIF 1136808076 M * Hollow vcontext -C -x 123 -f HIDE_NETIF -- vcontext ... 1136808099 M * Hollow the second command is vncontext of course 1136808166 M * NikDaPhreak Hollow: mmm, ok :-) I was too tired yesterday to test the options/flags... and today I must work... more or less.... ;-) 1136808181 M * Hollow too bad ;) 1136808197 M * NikDaPhreak Hollow: i'll test it later this evening... 1136808231 M * NikDaPhreak Hollow: I'm just doing it for living.. you know... ;-) 1136808357 Q * Andi Remote host closed the connection 1136808396 J * Andi ~Andi@socks.tuwien.ac.at 1136808594 M * Hollow NikDaPhreak: http://phpfi.com/95353 1136809084 M * NikDaPhreak Hollow: perfect :-) 1136809088 M * Hollow ;) 1136809109 M * Hollow commited the fix to svn 1136809232 M * NikDaPhreak Hollow: if I have a bit more time tonight, I might even rewrite the vserver script to honour the vncontext + settings in a separate (??) file 1136809328 M * Hollow i just started it, but it just a matter of some lines in scripts/lib/vps.sh 1136809376 M * NikDaPhreak Hollow: it might be a bit faster if you do it ;-) I can test/debug it later ;-) 1136809400 M * Hollow great :) 1136809474 M * Hollow hm, Bertl has not included the persistant patch for network contexts yet :/ 1136809477 M * Hollow won't work without it 1136809583 M * NikDaPhreak Hollow: if you call vncontext in the vserver script and not in the vps.sh? 1136809612 M * NikDaPhreak Hollow: or is there a separate patch for that? 1136809614 M * Hollow no, to make it work with the vserver command (which uses vps.sh) 1136809624 M * Hollow we introduced persistant contexts 1136809632 M * Hollow it makes life easier in the setup state 1136809640 M * Hollow but we need this for network contexts as well 1136809652 M * NikDaPhreak Hollow: yes, I studied the scripts yesterday ;-) I got the point 1136809680 Q * coocoon Ping timeout: 480 seconds 1136809687 M * NikDaPhreak Hollow: BTW I tried to play with PERSISTANT flag yesterday & assumed, I was too tired, 'cause it never worked.... 1136809729 M * Hollow NikDaPhreak: the thing here is, that in the old utils, all command were chained.. like vcontext .. -- vflags .. -- vsched etc 1136809750 M * Hollow but we want to unchain it, and if you just start the context and exit, the context died as well, because no process is left 1136809759 M * Hollow so we introduced persistant 1136809875 M * NikDaPhreak Hollow: i noticed that for contexts I played with your utils for two or 3 hrs... The idea is great. 1136809917 M * Hollow yeah, i hope there will be many more great ideas for the utils :) 1136810058 A * NikDaPhreak afk 1136810318 J * coocoon ~kvirc@p54A0750D.dip.t-dialin.net 1136810513 M * schellh hollow: vserver build was unable to cpmplete the build but it has not deletet the dirs it created 1136810519 M * schellh I: Validating libgcrypt11 1136810519 M * schellh E: Couldn't download libgcrypt7 1136810535 M * schellh ls /vservers/ 1136810536 M * schellh test 1136810589 M * schellh sorry cocoon told me the newest utils would delete failed stuff 1136810825 M * Hollow which utils? 1136810870 M * schellh util-vserver-0.30.209 1136810931 M * Hollow *shrug* i don*t think it removes orphaned files 1136810966 M * schellh ok so this was wrong.. 1136811001 M * coocoon hm i don't know I 've started with util-vserver30-208 1136811016 M * coocoon there I tried to install woody 1136811030 M * coocoon it works but doesn't start 1136811033 Q * balbir Quit: Leaving 1136811035 M * schellh *GG* 1136811046 M * schellh on a sparc woody will cause kernel OOps 1136811050 M * coocoon as i install with util-vserver 209 1136811053 M * Hollow sorry, i don't know anything about debian ;) 1136811071 M * coocoon after installation all files got erased 1136811101 M * coocoon sorry i do't know anything about sparc 1136811103 M * coocoon ;-) 1136811166 M * schellh hehe i am a looooser 1136811196 M * schellh in the debian channel i asked things about a debootstrap install of debian out of a running gentoo and they would harrass me because of gentoo 1136811209 M * schellh then i ask how i can create my own install cd for sparc and no one answers 1136811223 M * Hollow yeah, that's debian how i know it 1136811239 M * schellh maybe i should take solaris10 and their domains for my purposes. but i wouldnt like solaris 1136811288 M * Hollow maybe you should install gentoo guests instead of debian ones ;) 1136811304 M * coocoon or try it with ubuntu 1136811327 M * schellh ok ill just try and start some ubuntu installation for sparc 1136811376 M * schellh he they dont seem to support sparc 1136811398 M * Hollow schellh: you have gentoo vserver running on sparc? 1136811439 M * schellh i didnt try with gentoo guest but thats the last thing i will do when my debian efforts fail 1136811450 M * schellh but my host is gentoo yes 1136811471 M * Hollow hm, the thing is.. we don't have vserver stages for sparc 1136811493 M * schellh i guess i could use some ordinary level3 stage from gentoo sparc ? 1136811549 M * Hollow schellh: yeah, you could do that, but you need to install baselayout-vserver afterwards.. it's explained on top of http://linux-vserver.org/GentooGuestHowto how to do that 1136811556 M * coocoon the best how to for installing debian vserver http://deb.riseup.net/vserver/ 1136811574 M * Hollow you should probably do that in a chroot, and not start the vserver before 1136811970 M * Hollow NikDaPhreak: i added the network support.. http://dev.croup.de/proj/vserver-utils/changeset/24 1136812094 Q * id23 Quit: Leaving 1136812159 M * schellh is the verver proc unhide script within vservers-default or vservers-legacy ? 1136812889 M * NikDaPhreak Hollow: errr.... isn't it a better idea to have the net xid in a separate variable? imagine using the same net ctx for 2 or 3 vservers, which, for some reason, need the same net ctx.... 1136812914 M * NikDaPhreak Hollow: maybe with a default failover to VX_XID 1136812929 M * NikDaPhreak Hollow: makes the setup just a bit more flexible 1136812966 M * FaUl schellh: wy using woody? 1136812995 M * schellh no woody wont work, i just tried because i remebered a succesfull install on such a machine 1136813000 M * Hollow NikDaPhreak: indeed... 1136813031 J * Roey ~katz@h-69-3-4-130.mclnva23.covad.net 1136813040 M * Hollow but if you want to use one nx for multiple vx, it won't work anyway, because also in the setup state you can't nx_create muliple times 1136813064 M * NikDaPhreak Hollow: ok, UR right 1136813085 N * lchvdlog lchvdlch 1136813089 M * lchvdlch morning all 1136813095 M * NikDaPhreak Hollow: than it is maybe better to leave it this way... 1136813101 M * Hollow though the idea is nice, but it's very complex imo 1136813119 M * Hollow maybe we'll add it at some point in the future ;) 1136813143 M * NikDaPhreak Hollow: it can still be acchived by starting a vserver in an allready running net ctx... 1136813154 M * NikDaPhreak Hollow: this done by hand 1136813270 M * Hollow yeah could be done, but another thing is NGnet (or what is it will be), so we have to look at future network devlopment before implementing complex ideas imo 1136813290 M * NikDaPhreak Hollow: :-) 1136813299 M * schellh hi lchvdlog 1136813309 M * NikDaPhreak Hollow: it was not that complex on the first tought.... 1136813364 M * Hollow yeah, you could do it with the already running nx trick or so, but i tend not to make workarounds or hacks, but fixes and features ;) 1136813441 M * NikDaPhreak Hollow: :-) 1136813458 M * Hollow that's also why i started these new utils ;) 1136813474 M * Hollow so i'll keep an eye on legacy code and hacks ;) 1136813677 J * Johnnie ~john@acs-24-154-53-16.zoominternet.net 1136814479 M * Roey hello 1136814489 M * schellh hi 1136814490 M * Roey any new ideas on openvpn under vserver? 1136814492 M * Roey hey schellh 1136814570 M * FaUl schellh: i'm running sarge sucessfully on 2.6.14.4-vs2.1.0 (with some minor issues) 1136814585 M * FaUl on my dual-ultrasparc-II 1136814681 M * schellh well i could not install it because the install cd is lacking the sbus based scsi adapters i have 1136814693 M * schellh i would love to have a selfmade install cd but dont know howto .. 1136814730 M * schellh the url http://www.gentoo.org/doc/de/vserver-howto.xml tells abour some vserver-new script, where would i get that ? 1136814922 J * hwarrier ~harikb@c-67-164-93-186.hsd1.ca.comcast.net 1136815006 J * mef ~mef@targe.CS.Princeton.EDU 1136815257 M * FaUl no idea 1136815257 J * Smutje_ ~Smutje@xdsl-87-78-41-172.netcologne.de 1136815318 M * coocoon maybe in the stage 1136815356 M * coocoon Sie müssen einen als stage3 preparierten vserver von http://dev.gentoo.org/~hollow/vserver/stages herunterladen und das Skript wird Sie dann durch den Vorgang führen 1136815366 Q * Smutje Ping timeout: 480 seconds 1136815406 Q * hwarrier Ping timeout: 480 seconds 1136815521 Q * mef Quit: using sirc version 2.211+KSIRC/1.3.12 1136815635 Q * NikDaPhreak Quit: Hybernating my brain.... 1136815711 M * Hollow oh wow, they have translated it into german 1136815759 M * schellh but there is no vserver-new within the util-vserver-0.30.209 (?) 1136815772 M * FaUl Hollow: aren't you german as well? :-) 1136815794 M * coocoon there are also polish translation 1136815919 M * Hollow FaUl: yeah, but i hate german "fachliteratur" for computers, it always sounds horrible 1136815937 M * FaUl Hollow: yes, indeed :-) 1136815954 M * Hollow also no german locales here, bah bah 1136815976 M * FaUl Hollow: well, i havn't got my system repaired from removing all locales 1136816011 M * FaUl slrn and mutt have still problems with 'umlaute", any idea on that what or how to fix? 1136816072 M * schellh FauL : can you make me a new debian sarge guest ? when i build one, it wont be successful 1136816082 M * FaUl yes, i can 1136816112 M * schellh would be cool :) 1136816113 M * FaUl where did it stop? you have to touch the /etc/vservers/.defaults/nonamespace 1136816136 M * schellh I: Validating libgcrypt11 1136816136 M * schellh E: Couldn't download libgcrypt7 1136816152 M * FaUl else the vserver won't startup proberly 1136816160 M * FaUl strange, which debian-mirror have you chosen? 1136816213 M * schellh http://ftp.debian.de/debian 1136816254 M * schellh and the one i build last week succesfully didnt have any content within /etc 1136816265 M * schellh its kinda sucky anything 1136816419 M * FaUl hm, strange 1136816427 M * FaUl it's working fine at my vserver 1136816437 M * FaUl would you like a skeleton-vserver tared? 1136816443 M * schellh yes ! 1136816447 M * schellh i did a ARCH=sparc vserver test build -m debootstrap -n test --force -- -d sarge 1136816478 M * FaUl hm, sparc64, right? 1136816577 M * schellh yes 1136816615 M * schellh i believe i should be also able to take this working skeleton and install it as my host on a different partition with my own kernel or ? 1136816617 M * FaUl just taring the stuff together :-) 1136816636 M * FaUl uhm, don't know exactly 1136816648 M * coocoon sorry i have no experience with sparc but i have read that debootstrap will recognise automatically which architecture the system has 1136816650 M * FaUl its basically some minimal debian-installatino 1136816706 M * schellh no it wont find the ultrasparc-ii blackbird arch within the sparc debootstrap file 1136816748 M * FaUl well - on my ultra2 it works out of the box 1136816750 M * schellh and the debootstrap (debian only thing) which can be installed in gentoo - gentoo package wont work because it lacks some scripts 1136816754 M * kilian mmh, someone able to enlighten me what's going wrong here: 1136816755 M * kilian http://backend.verfaction.de/~kk/hppa-oops 1136816758 M * schellh kind of fucked up any approach i make 1136816770 M * kilian is a 2.6.15 on hppa 1136816833 M * coocoon hm do u use the latest debootstrap version 1136816846 M * schellh yes thats the only ones available on the ftps 1136816906 M * coocoon http://ftp.debian.de/debian/pool/main/d/debootstrap/ 1136816907 M * kilian to be precise on hppa non-smp parisc (i.e. 32bit) 1136816919 M * coocoon here are more i think so 1136816954 M * coocoon debootstrap_0.3.3.tar.gz i have used this one 1136816955 M * schellh cocoon: one for sparc and one old for sparc and debian woody 1136816970 M * coocoon this is for all architectures 1136817006 M * coocoon or isn't it 1136817022 M * schellh maybe 1136817027 M * coocoon it looks like this one debootstrap_0.3.3_all.deb 1136817030 M * coocoon but is tared 1136817084 M * schellh its with the right arch i suppose. doing it now 1136817201 M * coocoon u know i have no experience with sparc 1136817207 M * coocoon ;-) 1136817275 M * schellh *G still the error I: Validating libgcrypt11 1136817275 M * schellh E: Couldn't download libgcrypt7 1136817306 M * kilian schellh: you're trying to bootstrap a sid with the sarge debootstrap version? 1136817322 Q * Andi Remote host closed the connection 1136817349 M * schellh this is my command 1136817350 M * schellh ARCH=sparc vserver test build -m debootstrap -n test --force -- -d sarge 1136817357 M * schellh there is no sid bootstrap (?) 1136817384 M * kilian schellh: mmh, strange.. the sarge deboostrap run should be ok 1136817397 M * kilian schellh: but if you want i can get you a chroot tarball for sarge 1136817420 M * schellh FauL is doing some for me thanks 1136817427 M * schellh its sparc64 do you have this ? 1136817429 M * FaUl schellh: https://wiki.chaostreff-dortmund.de/skel.tar.bz2 1136817455 M * kilian schellh: there is no sparc64 userland for Debian AFAIK 1136817455 M * schellh :)) 1136817481 M * schellh whats a userland ? FauL dont you have a userland ?? *G 1136817505 M * kilian schellh: userland as opposed to kernel.. there's sparc64 kernels, but the "tools" are sparc, not sparc64 1136817532 M * kilian schellh: at least until multiarch is finally going live for Debian ;) 1136817541 M * schellh ic 1136817551 M * FaUl schellh: its a 32bit-sparc-userland - the sparc64-userland is actually slower than the 32bit exept for some very special purposes 1136817609 M * kilian FaUl: is there something like debian-ppc64.alioth for those who want to test at least? 1136817627 M * FaUl kilian: no idea,n ever used it 1136817658 M * FaUl but i know that the gentoo-folks patches even the makefiles to prevent building sparc64-stuff because it's slower then sparc32 :-) 1136817716 M * schellh the gentoo package baselayout-vserver seems to be x86 only again, i could cry 1136817728 M * kilian FaUl: not sure i do consider that "proof", but well.. ;) 1136817737 M * coocoon that is no problem 1136817766 M * kilian FaUl: i'd rather say the fact there's no debian-sparc64 even on alioth is good enough for me ;) 1136817767 M * coocoon there is a setting in the kernel which u can set for using x86 vservers 1136817780 M * FaUl schellh: just hit hollow, AFAIK he's maintaining it :-) 1136817787 M * coocoon or 32 bit architectures 1136817811 M * schellh cocoon : x86 guests on sparc ? really i dont think so ? 1136817822 M * FaUl haven't seen it though 1136817834 M * FaUl i don't think so either 1136817844 M * coocoon ok sorry i am only thinking in 64 buit 1136817846 M * coocoon sorry 1136817862 M * FaUl coocoon: you mean x86_64 1136817863 M * FaUl :-) 1136817870 M * coocoon yes 1136817873 M * coocoon thanx 1136817874 M * Hollow pong? 1136817902 M * Hollow schellh: it should compile on any arch which is supported by normal baselayout, but we don't have resources to test it 1136817935 M * FaUl Hollow: i may give you an account on that ultra1-box as well as bertl, but i'd to change the harddisk first 1136817960 M * Hollow FaUl: how long does a stage1 install take on this box? 1136817977 M * FaUl no idea, never tried this 1136817988 M * FaUl you may, but i guess a long time since it only got 166mhz 1136817996 M * Hollow oh ;) 1136818024 M * FaUl it's one of the first 64bit processor from sun - an ultrasparc1 :-) 1136818083 M * schellh well i have 3 ultras in the cellar should i bring up one to test ? 1136818111 M * kilian schellh: what kinda ultras? 1136818134 M * schellh i think 150-200mhz 1136818147 M * kilian usparcI oder usparcII? 1136818169 M * schellh i really dont know now 1136818212 M * kilian mmh, if it's usparcII they might be interesting, but AFAIK usparcI is quite boring to use as Debian box 1136818243 M * schellh we also have 2 enterprise 3500 and 2 enterprise 450 here .. but i cant give them away 1136818254 M * kilian *g* 1136818271 M * kilian you can leave 'em in place and we'll just use them right where they are if they have sufficient network access ;) 1136818280 M * schellh hehe 1136818341 M * kilian ;) 1136818527 M * schellh ok ill paste some stuff: 1136818528 M * schellh vserver test start 1136818528 M * schellh mount: none already mounted or . busy 1136818528 M * schellh /etc/vservers/test/fstab:1:1: failed to mount fstab-entry 1136818528 M * schellh mount: none already mounted or . busy 1136818528 M * schellh /etc/vservers/test/fstab:2:1: failed to mount fstab-entry 1136818528 M * schellh chbind: vc_set_ipv4root(): Invalid argument 1136818544 J * Doener doener@i5387F5F8.versanet.de 1136818549 M * schellh my fstab: 1136818550 M * schellh none /proc proc defaults 0 0 1136818550 M * schellh none /dev/pts devpts gid=5,mode=620 0 0 1136818620 J * DuckMaster ~duckx@195.75.27.158 1136818700 M * schellh i configured anything like on my x86 vservers, nodename interface etc 1136818722 M * schellh touched the nonamespace, set init mark to default 1136818837 J * PilatomiK ~tek@ADijon-151-1-113-73.w83-203.abo.wanadoo.fr 1136818855 M * schellh argh 1136818862 M * schellh ./testme.sh 1136818862 M * schellh Linux-VServer Test [V0.14] Copyright (C) 2003-2005 H.Poetzl 1136818862 M * schellh chcontext is working. 1136818862 M * schellh chbind: vc_set_ipv4root(): Invalid argument 1136818862 M * schellh chbind failed! 1136818885 M * schellh when i had my kernel compiled with the vserver api it was working! 1136818895 M * schellh i think ill reboot my old kernel 1136818909 M * schellh why do some gentoo docy tell to not do this !?? 1136818913 M * schellh docs 1136819061 M * coocoon from where do u know that this is the mistake 1136819099 M * schellh well there is some testme.sh which tests the kernel functions 1136819145 M * schellh http://www.gentoo.org/doc/de/vserver-howto.xml here is written to deactivate the legacy kernel api 1136819152 M * coocoon yes i know 1136819157 M * schellh this is the only change i did and now its not working anymore 1136819163 M * schellh ill reboot old kernel 1136820635 M * daniel_hozac schellh: are you running 2.1.0+ and disabled CONFIG_VSERVER_DYNAMIC_IDS? 1136820761 M * schellh my 2.10 patched kernel does not have such an option 1136820774 M * schellh all i changed was to disable CONFIG_VSERVER_LEGACY 1136820836 M * schellh ok seems with this option is CONFIG_VSERVER_DYNAMIC_IDS=y gone 1136820838 Q * FireEgl Ping timeout: 481 seconds 1136820975 M * daniel_hozac makes sense for it to depend on legacy. 1136821082 J * undefined ~undefined@adsl-68-93-109-94.dsl.rcsntx.swbell.net 1136821504 M * DuckMaster moin 1136821523 M * schellh hi 1136821550 M * DuckMaster Did anyone tried stable version 2.01 on linux 2.6.15 ? 1136821591 M * DuckMaster Well we gonna reboot on it tonight ... so I would like to know if there is some stuff I should be aware of ... :) 1136821601 N * Bertl_zZ Bertl 1136821606 M * Bertl evening folks! 1136821607 M * DuckMaster Moin bertl :) 1136821614 M * Bertl DuckMaster: yes I tried it :) 1136821616 M * coocoon hello 1136821621 M * coocoon bertl 1136821646 M * DuckMaster lol, shure you did ? :) Well if no problem or objection, I will test it tonight ;) 1136821664 M * DuckMaster It is compiled I just need to reboot the computer :) 1136821677 M * Bertl hey coocoon! 1136821697 M * DuckMaster Bertl, I will give you some feedback if everything goes fine or wrong .. 1136821748 M * Bertl excellent! 1136821867 M * coocoon which kind of error is thatchbind: execvp(): No such file or directory 1136821883 M * coocoon chbind: execvp(): No such file or directory 1136821904 M * kilian moin Bertl =) 1136821934 M * kilian Bertl: any good hint where http://backend.verfaction.de/~kk/hppa-oops might be coming from? 1136821947 M * kilian Bertl: that's hppa parisc up (i.e. 32bit kernel and 32bit userland) 1136822002 M * Bertl ah, the soft lockup thing ... 1136822029 M * Bertl seems that code is really new and not well tested, IIRC it also failed on the powerpc (sometimes) 1136822045 M * kilian Bertl: so that means? 1136822066 M * Bertl I would suspect some cache/memory coherence issues (out of the blue) 1136822083 M * Bertl at least that would explain why hppa and ppc64 sees it but x86_64 and i386 not 1136822110 M * Bertl kilian: for you that probably means to disable it for now 1136822123 M * Bertl (or figure what goes wrong there) 1136822187 M * Bertl coocoon: depends, but looks like a binary or library cannot be found ... 1136822196 M * kilian Bertl: uhm, where do i disable that? 1136822212 M * Bertl kilian: it's a kernel option, brand new (detect soft lockups) 1136822256 M * kilian CONFIG_DETECT_SOFTLOCKUP=y 1136822258 M * kilian that one? 1136822271 M * Bertl yep, precisely 1136822321 M * Bertl I had a look at the code, looked fine so far, but as I said, probably cache/memory coherence ... 1136822338 M * Bertl (or some kind of 'design' flaw for SMP :) 1136822339 M * kilian Bertl: ok, will give that a spin =) 1136822355 M * kilian Bertl: that hppa box is no SMP ... 1136822397 M * Bertl ah, okay, that is interesting then ... 1136822407 M * Bertl let me take another look at the relevant code 1136822409 M * kilian Bertl: yep, and the kernel is not SMP either 1136822417 M * Bertl even better 1136822425 M * kilian *g* 1136822439 M * Bertl means I don't have to check spinlocks either :) 1136822625 J * FireEgl Atlantica@Atlantica.US 1136822834 M * Bertl welcome FireEgl! 1136823358 M * kilian Bertl: alright.. in some ..ehrm.. hours.. we'll know ;) 1136823424 M * Bertl hmm, why hours? 1136823515 M * kilian because i still need to find out how this new debian kernel build system works.. and apparently it has just purged the build directory ;) 1136823538 M * Bertl lol 1136823540 M * kilian Bertl: thus i'll need to have the kernel compiled from scratch.. and ehrm, this hppa is not really "performant" ;) 1136823553 Q * shedi Quit: Leaving 1136823555 M * Bertl why not cross compile? 1136823570 M * kilian mmh, i guess that would take just as long :-p 1136823590 M * Bertl depends on the machine ... 1136823595 M * kilian (for me to read into it and setup the whole thing) 1136823612 M * kilian but i'll look into that after going shopping and stuff 1136823892 M * coocoon can someone tell me how to fix this 1136823894 M * coocoon chbind: execvp(): No such file or directory 1136823938 M * Bertl coocoon: a) what tools do you have (best upload the output of 'vserver-info - SYSINFO' somewhere, e.g. pastebin) 1136823952 M * Bertl b) when do you get that message? 1136824148 M * coocoon a) util-vserver: 0.30.209 kernel 2.6.14.3-vs2.1.0-rc10 b) starting the v.... scripts 1136824238 M * Bertl okay, any chance to get a peek at the SYSINFO? and what v.... scripts? 1136824271 M * coocoon Kernel: 2.6.14.3-vs2.1.0-rc10 1136824271 M * coocoon VS-API: 0x000100ff 1136824271 M * coocoon util-vserver: 0.30.209; Jan 6 2006, 20:23:38 1136824271 M * coocoon Features: 1136824272 M * coocoon CC: gcc, gcc (GCC) 3.3.5 (Debian 1:3.3.5-13) 1136824272 M * coocoon CXX: g++, g++ (GCC) 3.3.5 (Debian 1:3.3.5-13) 1136824274 M * coocoon CPPFLAGS: '' 1136824274 M * coocoon CFLAGS: '-g -O2 -std=c99 -Wall -pedantic -W' 1136824275 M * Bertl the output of testme.sh would be interesting too, btw ... 1136824276 M * coocoon CXXFLAGS: '-g -O2 -ansi -Wall -pedantic -W -fmessage-length=0' 1136824276 M * coocoon build/host: i686-pc-linux-gnu/i686-pc-linux-gnu 1136824278 M * coocoon Use dietlibc: yes 1136824278 M * coocoon Build C++ programs: yes 1136824280 M * coocoon Build C99 programs: yes 1136824280 M * coocoon Available APIs: compat,v11,v13,fscompat,net,oldproc,olduts 1136824282 M * coocoon ext2fs Source: e2fsprogs 1136824282 M * coocoon syscall(2) invocation: alternative 1136824284 M * coocoon vserver(2) syscall#: 273/glibc 1136824286 M * coocoon Paths: 1136824286 M * coocoon prefix: /usr 1136824288 M * coocoon sysconf-Directory: /etc 1136824288 M * coocoon cfg-Directory: /etc/vservers 1136824290 M * coocoon initrd-Directory: $(sysconfdir)/init.d 1136824290 M * coocoon pkgstate-Directory: /var/run/vservers 1136824292 M * coocoon vserver-Rootdir: /srv/vservers 1136824326 M * coocoon chcontext is working. 1136824326 M * coocoon chbind is working. 1136824326 M * coocoon Linux 2.6.14.3-vs2.1.0-rc10 i686/0.30.209/0.30.209 [Ea] (0) 1136824326 M * coocoon VCI: 0002:0001 273 031101f6 1136824326 M * coocoon --- 1136824328 M * coocoon [000]# succeeded. 1136824328 M * coocoon [001]# succeeded. 1136824330 M * coocoon [011]# succeeded. 1136824330 M * coocoon [031]# succeeded. 1136824332 M * coocoon [101]# succeeded. 1136824332 M * coocoon [102]# succeeded. 1136824334 M * coocoon [201]# succeeded. 1136824334 M * coocoon [202]# succeeded. 1136824343 M * daniel_hozac http://pastebin.com/ please! 1136824360 M * Bertl coocoon: okay, looks good, so which scripts do you give the message? 1136824361 M * coocoon b) v_named v_sshd v_xinetd 1136824374 M * Bertl hmm, are they part of 0.30.209, sec ... 1136824405 M * coocoon first i started them no error messages only v_gated 1136824411 M * coocoon shows the error message 1136824417 M * coocoon after reboot 1136824422 M * coocoon all shows it 1136824422 M * daniel_hozac why are you using them at all? 1136824432 M * coocoon ok pastebin next 1136824466 M * Bertl coocoon: when do they show it? 1136824475 M * coocoon only at starting 1136824489 M * coocoon hm i wanted to try to have access to my vserver 1136824516 M * coocoon today have tried something with nat got some rules from phreak 1136824521 M * Bertl let's try one, like this: 1136824537 M * coocoon my vswerver has access to the internet but i can't achieve it with my ssh client 1136824571 M * Bertl 'bash -x /etc/init.d/v_named restart' 1136824635 M * coocoon + USR_LIB_VSERVER=/usr/lib/util-vserver + exec /usr/lib/util-vserver/vsysvwrapper named restar exec /usr/sbin/chbind --ip eth0 /etc/init.d/named restar 1136824635 M * coocoon ipv4root is now hostip chbind: execvp(): No such file or directory 1136824699 M * Bertl first, it looks like you tryped restar (but that doesn't matter) 1136824705 M * Bertl *typed even 1136824724 M * coocoon it is my shell 1136824725 M * Bertl then, it very much looks like that there is no IP assigned for the host 1136824725 J * cdrx ~legoater@cap31-3-82-227-199-249.fbx.proxad.net 1136824735 M * coocoon i have changed the host ip 1136824740 M * coocoon SECRET 1136824743 M * Bertl welcome cdrx! 1136824747 M * coocoon set hostip 1136824751 M * cdrx eh hi ! 1136824780 M * Bertl coocoon: okay, what about /etc/init.d/named, does it exist? 1136824905 M * coocoon no not there 1136824915 Q * Milf Ping timeout: 480 seconds 1136824915 M * coocoon there it is/usr/sbin/named 1136824930 M * Bertl so, how should it work then, if there is no runlevel script for named? 1136824953 M * Bertl where is your runlevel script for named? 1136824956 M * coocoon 4885 0.0 0.1 38640 2348 ? Ss 17:42 0:00 /usr/sbin/named -c /etc/bind/named.conf 1136824968 M * coocoon is it there false 1136824974 M * Bertl yeah, that's fine, that's the resolver 1136824975 M * coocoon this is from ps aux 1136824988 M * coocoon starting the nam server works 1136824991 J * multiple null@62.162.237.139 1136825007 M * Bertl but /etc/init.d usually contains scripts (like named) for starting the services, where are they on your machine? 1136825011 M * Bertl welcome multiple! 1136825018 M * coocoon i don't know 1136825023 M * multiple hi 1136825027 M * coocoon i only have there bind9 1136825043 M * Bertl coocoon: so you want to use/create v_bind9 then 1136825047 M * multiple What is vserver? 1136825062 M * coocoon if i must 1136825064 M * coocoon i will 1136825133 M * Bertl multiple: A soft partitioning concept based on Security Contexts which permits the creation of many independent Virtual Private Servers (VPS) that run simultaneously on a single physical server at full speed, efficiently sharing hardware resources. 1136825181 M * coocoon is this responsible for that, that i can't connect from outside to my vserver 1136825183 M * coocoon over ssh 1136825185 M * Bertl coocoon: well, you can do a bunch of other things too, but the v_* scripts are just wrappers for existing scripts, so if you are missing the named script, the v_named script will not succeed to call the other one 1136825206 M * coocoon ok i understand 1136825235 M * Bertl coocoon: and no, the ssh conenct just depends on two things, a) that your host's ssh does not bind the IP of the guest, and b) the the guest is actually reachable and has sshd running 1136825241 M * coocoon i don't need them but i thought, when i start them it will solve my problem 1136825276 M * Bertl coocoon: the best approach is to write the host-only IP(s) into the sshd config file (as Listen address) 1136825286 M * coocoon that i have done 1136825304 M * coocoon i have connect to the internet 1136825310 M * coocoon but can't access it 1136825319 M * coocoon my vserver from outsiode 1136825342 M * coocoon it looks very easy to read all docus 1136825351 M * coocoon but there are no further information 1136825388 M * coocoon I thought i depends on my firewall settings but also natting the vserver and the port to the vservers sshd 1136825409 M * coocoon conf didn't solve this 1136825444 M * coocoon I have set into dns the vhost00.hostname.net 1136825460 M * coocoon and also if make a ping on it my host ip will return 1136825750 M * coocoon no more answers 1136825770 M * coocoon maybe i must setupo the whole system new without configuring a firewall 1136825831 J * stefani ~stefani@superquan.apl.washington.edu 1136825934 J * ichigo ~m@203.81.238.222 1136825956 M * ichigo hello everybody. 1136826083 M * ichigo i have this problem. i installed vserver and succesfully created a ubuntu vserver from the ubuntu cd using debootstrap. but i can't start it it. i get this error. http://pastebin.com/497919 1136826137 M * Bertl welcome stefani! ichigo! 1136826180 M * stefani hola 1136826215 M * Bertl ichigo: kernel patch 2.1.0? 1136826259 M * Hollow heya Bertl 1136826262 M * ichigo i used patch-2.6.14.4-vs2.1.0.diff 1136826276 M * Bertl ichigo: k, you disabled legacy support 1136826284 M * Hollow Bertl: the persistant flag for nx is still missing in 2.1.0 1136826318 M * Bertl Hollow: hmm, okay, will add it ... 1136826323 M * Hollow thx 1136826327 M * ichigo so whats the fix?? do i recompile the kernel..? 1136826358 M * Bertl ichigo: well, your kernel does not provide the legacy interface the tools expect, so yes ... reconfigure and recompile/install 1136826393 M * Bertl ichigo: alternatively you can test with Hollow's tools 1136826400 J * jeeves ~Bob@c-24-11-171-10.hsd1.mi.comcast.net 1136826407 M * Bertl welcome jeeves! 1136826417 M * jeeves hey there, got good news....... 1136826421 M * Hollow Bertl: i added network support for the tools to svn today 1136826432 M * Bertl Hollow: excellent! 1136826441 M * Bertl jeeves: let's hear! 1136826496 M * jeeves My domestic CFO (wife) is allowing me to spend some money on a computer. I am going to replace my MB and CPU with a dual core Opteron. I am going to get two 19" monitors also. 1136826521 M * jeeves Are there issues with vserver andOpterons? 1136826545 M * ichigo what options should i enable. here's part of the config file that relates to vserver http://pastebin.com/497929. what should i change. 1136826651 M * Bertl jeeves: not that we know of ... hmm, except sometimes performance is too high :) 1136826669 M * Bertl # 1136826671 M * Bertl CONFIG_VSERVER_NGNET=y 1136826682 M * Bertl ichigo: that's the culprit here 1136826699 M * Bertl ichigo: i.e you want CONFIG_VSERVER_LEGACYNET=y isntead 1136826699 M * FaUl Bertl: how far is ngnet right now? 1136826730 M * Bertl FaUl: nothing changed in ngnet since the loopback isolation proof of concept 1136826749 M * _are_ jeeves: I sucessfully run vserver on a dual-opteron box 1136826762 M * FaUl Bertl: so there isn't any usable code right now, right? 1136826777 M * ichigo thanks bertl. will fix that. 1136826781 M * jeeves _are_: sweet 1136826810 M * _are_ 64-bit kernel, not 32-bit, but 32-bit-emulation enabled in kernel. 1136826819 M * _are_ (and some 32-bit vservers running) 1136826836 M * jeeves _are_: If I just install the _64 kernel on the box before upgrading, will I have to reinstall anything else? 1136826876 M * _are_ yes 1136826880 M * Bertl ichigo: you're welcome! btw, you probably don't want/need the XID_TAG_NFS and PROPAGATE (not sure if you want the HISTORY debug stuff either) 1136826893 M * _are_ actually: no 1136826903 M * jeeves I was hoping I didn't need to reinstall everything on it. 1136826916 M * _are_ the kernel can run both, the libraries/code decides what you actually run 1136826950 M * ichigo Bertl. there is no CONFIG_VSERVER_LEGACYNET entry in the config file. is that the problem. do i just add it. 1136826976 M * _are_ and unlike some other architectures that have a lib32 and lib64 environment, x86_64 at least on debian has only one /lib, easiest way to run both is having different chroot/vservers with the other libraries. 1136827030 M * _are_ if you are able to run 64 and/or 32bit is decided by what compile options you used for the kernel. However, I run 64-bit native and have a few 32bit vservers and I am not to sure the opposite would work. 1136827039 M * Bertl ichigo: just remove the NGNET entry, and do 'make oldconfig' then answer the questions properly 1136827074 M * jeeves I am not too worried about running both, I just didn't want to reinstall everything. I was hoping to pull the hard drive and put it in the new box and have it boot up with the _64 kernel installed. 1136827095 M * Bertl jeeves: should work 1136827111 M * jeeves Sweet!!! 1136827149 M * jeeves The downside is the Domestic CFO won't let me to it till mid Febuary. 1136827249 J * oliwel ~mail-at-o@host-62-245-151-178.customer.m-online.net 1136827263 M * coocoon thanx for all help 1136827264 M * Bertl hey oliwel! :) 1136827265 M * coocoon bye 1136827275 M * Bertl coocoon: cya! 1136827279 Q * coocoon Quit: KVIrc 3.2.0 'Realia' 1136827343 M * oliwel Bertl: Ring :) 1136827395 M * Bertl soo .. kernel version? 1136827410 M * oliwel *g* 1136827414 M * oliwel two of them 1136827422 M * oliwel 2.6.14-vs2.0.1-gentoo 1136827424 M * oliwel 2.6.13-vs2.1.0-pre5-gentoo 1136827441 M * oliwel these are the both peers of the drbd system and I have probs on both... 1136827450 M * Bertl hmm, okay, let's look at the 2.0.1 first ... 1136827464 M * oliwel util-vserver: 0.30.209; 1136827467 M * oliwel ok 1136827486 M * oliwel so, as already in the mail - vnamespace -e doesnt work - no such process 1136827506 M * Bertl hmm, no, IIRC, you used the guest name, not the xid there :) 1136827570 M * oliwel tried both 1136827588 M * Bertl well, let's try again, first, I need 'vserver-stat' 1136827605 M * Bertl (please upload everything above 3 lines to pastebin.com) 1136827624 M * oliwel mom - my excuses - kernel is a 2.6.13-vs2.1.0-pre5-gentoo also...was on the wrong box 1136827631 M * oliwel ok mom 1136827670 M * Bertl Hollow: 2.6.13-vs2.1.0-pre5 do you have some details? 1136827703 Q * _are_ Quit: bbl 1136827737 M * oliwel http://pastebin.de/3626 1136827743 Q * DuckMaster Quit: Leaving 1136827750 M * Hollow Bertl: gentoo details? 1136827769 M * Bertl well, patches, fixes after that? 1136827815 M * Hollow hm, have to look 1136827822 M * Bertl Hollow: tx 1136827851 M * oliwel Bertl: so, what now ? 1136827857 M * Bertl oliwel: okay, what does 'vnamespace --enter 10069 -- cat /proc/mounts 1136827861 M * Bertl give? 1136827906 M * oliwel lot of mounts that should be there but nothing that shouldnt bee :) 1136827936 M * oliwel FYI: the suspected vserver was the last one started and was running for only some minutes 1136827938 M * Hollow Bertl: http://distfiles.gentoo.org/distfiles/genpatches-2.6.13-1.base.tar.bz2 and http://distfiles.gentoo.org/distfiles/genpatches-2.6.13-1.extras.tar.bz2 and the vserver patch 1136827945 M * oliwel so no restart of the other ones was done in between 1136828092 Q * jazzanova Quit: Connection reset by pear 1136828133 M * Bertl oliwel: well, I probably have to look at all the mounts of all three guests first (and the host) to get an idea ... 1136828147 Q * ichigo Quit: Leaving 1136828168 M * oliwel I promise there is nothing suspicous....but of course I can upload it.... 1136828174 M * Bertl oliwel: but you could also test with a recent 2.1.0 release and 2.0.1 to narrow down the issue, if you prefer ... 1136828198 M * oliwel so its a 2.6.13-vs2.1.0-pre5-gentoo 1136828222 M * oliwel but I must do a new kernel thes days for oterh reasons either...so if that helps... 1136828268 M * oliwel So here is the foirst mount http://pastebin.de/3627 1136828308 M * oliwel the other two are the same only differing in the drdb device monuted... 1136828343 M * Bertl okay, well, they for sure have a reference to the drbd now, right? 1136828395 M * oliwel no this is another drbd 1136828401 M * oliwel I have one of them for every vserver 1136828401 M * Bertl okay, let's talk about 'the issue' .. what do you expect to happen? what happens instead? 1136828428 M * oliwel ok, again :) 1136828437 M * oliwel I have ONE drbd device per vServer 1136828445 M * Bertl okay 1136828451 M * oliwel the device is mounted from the fstab of the server 1136828461 M * Bertl server means guest here? 1136828464 M * oliwel Yes 1136828470 M * Bertl okay 1136828482 M * oliwel So, as fas as I can see, everything is fine 1136828499 M * oliwel like you see on the pastebin, there is one drbd inside the server 1136828522 M * oliwel this is what I expect 1136828534 M * Bertl yep, mounted twice, because of the rbind mount 1136828537 M * oliwel when I shutdown the vserver , the drbd is unmounted/free'd 1136828552 M * Bertl if all processes are killed, yes 1136828556 M * oliwel Yes the first line is the device mount, the rest are bind's 1136828575 M * oliwel So, I want to "shutdown" the DRBD device 1136828598 M * oliwel which is possible only if there are no monuts on it 1136828604 M * Bertl right 1136828615 M * oliwel and drbd tools complain about "someone has openen the device" 1136828631 M * oliwel but I dont have an idea who/whatever should be "someone" 1136828640 M * Bertl and you know that all processes are gone (which belonged to the guest) 1136828671 M * oliwel Yes, but I cant see any process with "vps ax", vserver-stat 1136828677 M * Bertl i.e. no zombies or device processes hanging around, no helper or vnamespace enter there? 1136828694 M * oliwel hm, what are "helpers 1136828712 M * oliwel I alays looked for sometging within the killed namespace 1136828733 M * Bertl in earlier versions we had the vshelper hanging around in certain situations 1136828756 M * oliwel vps ax | grep vshelper is empty 1136828778 M * Bertl okay, I suggest the following: 1136828823 M * Bertl - get a 2.6.14.5 or 2.6.15 kernel with 2.1.0.3 1136828838 M * Bertl - compile it and keep the built tree around (for small modifications) 1136828870 M * Bertl - try a mount/umount (after reboot) of a single drb volume on the host 1136828887 M * Bertl (just to verify that this works as expected) 1136828896 M * Hollow Bertl: ah btw... what's the difference between 2.1.0 and 2.1.0.3? only the fairsched? 1136828905 M * oliwel uuuhh...this is my production system...would be a bit of work to migrate all the stuff 1136828933 M * Bertl oliwel: hmm, well, 2.1.0 is development ... 1136828947 M * oliwel Bertl: Actually, when I shutdown one of the other guests on the system, drbd is working fine - SAME MASCHINE !! 1136828957 M * Bertl Hollow: http://vserver.13thfloor.at/Devel/PAT-2.1.1/ 1136828959 M * oliwel Bertl: I like the bleeding edge :) 1136828971 M * Hollow i see 1136828977 M * Bertl oliwel: which one? 1136828991 M * oliwel Bertl: explain.... 1136829005 M * Bertl oliwel: which guest do you shut down to make it work? 1136829032 M * oliwel I tried the "www1" - I can shut it down and stop the drbd without any problems 1136829082 M * oliwel the wwwecom works fine too 1136829125 M * oliwel and the thris one works too 1136829144 M * Bertl so what does not work again? 1136829165 M * oliwel I have one DRBD device from a currently not running vserver 1136829171 M * oliwel I cannot shutdown the device 1136829182 M * oliwel but I dont see anything that access'es it 1136829213 M * oliwel So you see, the problem is not really reproducible 1136829247 M * Bertl hmm, well, maybe you experimented with vnamespace and still have some shell/process inside that namespace? 1136829309 M * oliwel i think I had some problems with shutting down the vserver 1136829318 M * oliwel but I cant see any namespaces 1136829332 M * Bertl well, they are hard to see :) 1136829347 M * Bertl check for processes in 'D' state or as I said, zombies 1136829349 M * oliwel hard or invisble 1136829354 M * oliwel vps ax ? 1136829375 M * Bertl vps auxwww 1136829386 M * Hollow what about vcontext --migrate --xid 1 -- lsof /dev/drbd/foo 1136829392 M * Hollow ? 1136829402 M * oliwel will this crahs my root contecxt :) 1136829422 M * Hollow no it just shows who uses /dev/drbd/foo 1136829469 M * oliwel vcontext --migrate --xid 1 -- lsof /dev/drbd/www2 1136829470 M * oliwel lsof: WARNING: can't stat() usbfs file system /proc/bus/usb 1136829472 M * oliwel Output information may be incomplete. 1136829482 M * oliwel Bertl: No "D" in the process list 1136829541 M * Bertl what problems did you have on shutdown? 1136829565 M * oliwel Shutdown of the vserver ? 1136829572 M * Bertl yup 1136829590 M * oliwel uhh - I had an error on shuting down a daemon... 1136829611 M * oliwel so nothing related to the vserver ittself 1136829628 M * oliwel I cant really remember but it might be that I ran into a "wait" situation 1136829635 M * Bertl what xid does this context have? 1136829641 M * oliwel 1207 1136829683 M * Bertl does vps auxwww show any process in that context? (unlikely) 1136829688 M * oliwel no 1136829760 M * oliwel BRB 1136829844 M * Bertl Hollow: do you have the 2.6.13-vs2.1.0-pre5-gentoo sources at hand? 1136829879 M * Hollow Bertl: no, sorry, but you should be able to create it with the patch tarballs above 1136829906 Q * Roey Quit: Leaving 1136829922 M * Hollow maybe oliwel can upload vserver-sources-2.1.0_pre5.tar.bz2 for you, so you don't have to adapt the vserver patch to genpatches 1136829929 M * Hollow though it's just a hunk in sysctl.h 1136829929 M * oliwel Bertl: I can offer you a tar of my src tree 1136829948 M * Bertl oliwel: hmm, no, let's check something in the source 1136829955 M * oliwel ok 1136829993 P * PilatomiK Leaving 1136830002 M * Bertl kernel/vserver/network.c please copy/paste (pastebin) the function dev_in_nx_info() 1136830160 M * oliwel grml - pastebin.de seems to not like me...cant post...you you have another service at hand ? 1136830184 M * Bertl pastebin.com 1136830226 M * oliwel http://pastebin.com/498031 1136830269 M * Bertl hmm, interesting, seems that was fixed in 2.0.1 but not 2.1.0 ... 1136830292 M * Bertl okay, expect a new 2.1.x release shortly 1136830300 M * oliwel LOL 1136830305 M * oliwel so there is a bug in the code ? 1136830307 M * Hollow don't forget persistant nx please :) 1136830319 M * Bertl oliwel: yes, there is definitely a bug :) 1136830331 M * oliwel something that affects my issue ? 1136830332 M * Bertl Hollow: okay, should not be too hard 1136830337 M * Hollow yeah :) 1136830340 M * oliwel some worksarounds ? 1136830351 M * Bertl oliwel: yes, otherwise I would not have checked it ... 1136830353 M * oliwel Guys, that costs you at least a beer :) 1136830366 M * oliwel Spend 4 hours yesterday on the issue.... 1136830376 M * oliwel *kidding of course* 1136830379 M * Bertl oliwel: just do not use networking, it doesn't happen if there is no networking :) 1136830408 M * oliwel Ah ok, so I will tell my customers that the servers are runnig fine, but they cant see there webpages 1136830435 M * Bertl yeah, I guess they will understand it ... 1136830438 M * oliwel I will make a "tag der offenen tür" so everybody can look at the pages in the rack-area ::) 1136830463 M * oliwel So, Hollow - when can I expect a new ebuild 1136830485 M * Hollow after the release 1136830496 M * oliwel in days ? Bertl ? 1136830512 M * oliwel So for me the question is waiting for the new release or going to 2.0... 1136830531 M * Bertl give me a minute, the patch should be similar to 2.0.1 1136830553 M * oliwel hanigng on, its ok when its in the next days....but I cant wait weeks 1136830556 M * Bertl http://vserver.13thfloor.at/Devel/PAT-2.0.1/delta-indev-fix01.diff 1136830589 M * Bertl seems to apply fine, I assume it will work too, testing now :) 1136830614 M * pflanze Bertl: have you seen my question?: 'do you have an idea why all processes of one of my vservers are at nice level -5 for no apparent reason?' 1136830631 M * Bertl yeah, I saw it .. my ideas to this are: 1136830639 M * pflanze Maybe it's because of restarts with init type=plain. 1136830645 M * Bertl - the nice value is specified in the config tree/file 1136830657 M * Bertl - the process starting the guest is at nice -5 1136830672 M * Bertl - the guest has too many capabilities and renices init 1136830702 M * pflanze ok I'll try those in succession 1136830704 M * oliwel the patch apples in contet.c but not in network.... :( 1136830740 M * Bertl hmm, applied fine here on 2.6.15-vs2.1.0.3 ... 1136830780 M * oliwel hmm - hunk fails on 2.6,13 1136830786 M * Bertl try again with -l 1136830829 M * oliwel l like "Ludwig" ? 1136830837 M * Bertl yes, it ignores spaces 1136830839 M * oliwel Hunk #1 FAILED at 408. 1136830841 M * oliwel 1 out of 1 hunk FAILED -- saving rejects to file kernel/vserver/network.c.rej 1136830851 M * Bertl so patch -p1 -l --dry-run see http://pastebin.com/498054 1136831222 M * Bertl Andi: which version is that? 1136831253 M * Bertl Andi: patch version and what arch are you on? ppc? 1136831257 M * Andi patch-2.6.15-vs2.0.1.2.diff and patch-2.6.15-vs2.1.0.3.diff too 1136831294 M * Andi Bertl: It's a IA32 1136831315 M * Bertl interesting, must be a configuration detail, let me check that ... 1136831329 M * Andi If I comment out the line 193 in mm/filemap_xip.c compilation works fine 1136831370 M * Bertl yeah, well, if you comment out everyting, compilation is always fine, just results are different :) 1136831373 M * Andi I've used my working vserver-config from 2.6.14.3... 1136831375 M * daniel_hozac #include missing? 1136831385 M * Bertl daniel_hozac: very likely ... 1136831407 M * Andi ok, where to add the include? in filemap_xip ? 1136831425 M * Bertl #include 1136831436 M * Bertl +#include 1136831440 M * Bertl #include 1136831458 M * Bertl and it's the execute in place functionality 1136831514 M * Andi ok, I'll try it... rmap and tlbflush are already included... 1136831546 M * Bertl yeah, that was the context, the line with the '+' was the addition :) 1136831565 M * Andi *oops* ;-) 1136831606 M * Andi ok, make-kpkg is running :-) 1136831697 Q * prae Quit: Execute Order 69 ! 1136832106 M * Andi thank you too -> no it works! 1136832124 M * Andi Bertl: so the patch needs an patch (-: 1136832133 M * Andi -n 1136832161 Q * Doener Ping timeout: 480 seconds 1136832173 M * Andi ok, I should go to bed... thank you two *g* 1136832201 J * Doener doener@i5387FCD1.versanet.de 1136832312 M * Bertl Andi: it will be fixed in the next release 1136832331 Q * FireEgl Read error: Connection reset by peer 1136832491 M * Andi Bertl: that's great :-) 1136834515 M * oliwel Bertl: Patch works fine against current 2.6.15-gentoo kernel, compiling currently and will reboot aroud midnight...will report here if you are there or via ML 1136834601 M * Bertl okay, make that, you can leave a message here even when I'm not there, the channel is logged and I 'try' to read up on the backlog ... 1136834724 M * Bertl okay, off for a while .. dinner and so ... back later ... 1136834727 M * oliwel Bertl: ok fine...will go out now and try my best when I am back 1136834733 N * Bertl Bertl_oO 1136834736 Q * oliwel Quit: Chatzilla 0.9.69.1 [Firefox 1.5/2005111116] 1136834741 M * harry cya Bertl_oO 1136835002 J * prae ~benjamin@sherpadown.net 1136835122 M * michal_ http://foxtrot.com/ 1136836147 J * shedi ~siggi@inferno.lhi.is 1136836162 J * FireEgl ~FireEgl@Atlantica.US 1136836683 J * bonbons ~bonbons@83.222.39.249 1136836779 J * _are_ ~are@62.112.159.81 1136836827 M * bonbons Hi, does someone know where/how util-vserver chooses output target of init.d-script's echoing? 1136837217 P * eugenesan 1136837727 M * daniel_hozac bonbons: the guest's initscripts? 1136837827 M * bonbons daniel_hozac: I don't think so, the guest has just plain standard init scripts and init itself outputs some text. Must be on util-vserver side when it starts the (dummy) init process 1136839093 J * Roey ~katz@h-69-3-4-130.mclnva23.covad.net 1136839101 M * Roey hi 1136839250 M * Roey How do I close off ports for vserver guests via the host's iptables? 1136839312 Q * Andi Quit: Chatzilla 0.9.68.5.1 [Firefox 1.0.7/20050922] 1136839338 M * bonbons Roey: Use the same rules as for ports on the host, just using the IP address of the appropriate guest 1136839356 M * Roey oh, ok. 1136839376 M * Roey bonbons: and this is done on the host yes? Because the guests presumably have no acccess to in-kernel structures. 1136839406 M * bonbons Roey: Yes, this must be done from the host 1136839412 M * Roey alright then 1136839414 M * Roey makes sense 1136839419 M * Roey thank you 1136840375 M * daniel_hozac bonbons: if you're using the sysv init style, /etc/vservers//tty (IIRC, if it exists) or stdin/stdout. 1136840436 M * daniel_hozac bonbons: for the plain init style, i don't think the output is available at all, the guest (should) lack /dev/console 1136840470 M * bonbons daniel_hozac Ok, will give it a try. My aim is to let a given guest start on a given tty (and output boot-process on it) 1136840802 M * bonbons daniel_hozac: just creating a device console with major and minor of the target tty does the job, thanks to the hint! 1136841420 M * Roey anyone here using FreeNX with VServer? 1136842198 Q * pflanze Quit: night 1136843509 Q * cdrx Quit: Leaving 1136843776 M * derjohn Roey, what is FreeNIX ? 1136843811 M * daniel_hozac google ;) 1136843828 M * daniel_hozac (without the I, of course) 1136843829 M * derjohn daniel_hozac, thanks for the tipp ! :) 1136843890 M * Roey derjohn: FreeNX is this really really fast X server implementation geared for remote usage. 1136843945 M * derjohn Roey, I already google-ed :) Sounds interesting. So you gonna create a remote desktop for your users? 1136844001 M * Roey derjohn: well I just wanna first try and make a session at home that I can access at work--as X is very, very slow over SSH. 1136844027 M * daniel_hozac ssh -C ;) 1136844043 M * daniel_hozac it's still slow, but it's a _lot_ better. 1136844099 M * Roey I tried 1136844105 M * derjohn currently I would also love to have a fast reliable remote desktop .... (external project work ... ) 1136844106 M * Roey I tried, I tried. It's all the same garbage 1136844109 M * Roey it's so slow. 1136844116 M * derjohn Roey, VNC ? 1136844117 M * Roey derjohn: check out FreeNX. 1136844124 M * Roey derjohn: nono, that is just a screenscraper 1136844129 M * derjohn Roey, but not in a guets yet? 1136844137 M * derjohn s/guets/guest/ 1136844140 M * Roey derjohn: FreeNX preserves X's semantics 1136844160 M * Roey derjohn: so that instead of scraping a screen, you're giving the server a commnad to draw a rectangle. 1136844161 M * Roey Get it? 1136844165 M * derjohn Roey, so: a client in a guest? 1136844173 M * Roey a client that holds a server. 1136844177 M * Roey I wanna know if that's possible. 1136844178 M * Roey like 1136844183 M * Roey openvpn you cna't use with openvpn clients. 1136844185 M * Roey er 1136844188 M * Roey vserver clients 1136844189 M * Roey sorry 1136844190 M * Roey not openvpn 1136844192 M * daniel_hozac sure you can. 1136844210 M * daniel_hozac even Enrico does that. 1136844213 M * derjohn Roey, foo? I meant an X Client in a vserver guest? 1136844236 M * Roey I mean an X client in a vserver 'guest', right. 1136844239 M * Roey daniel_hozac: neat 1136844293 M * daniel_hozac Roey: easiest way to find out is to try ;) 1136844302 M * derjohn Roey, sounds not so difficuelt as the hardware access is done be user user, i.e. the x server which is not the vserver 1136844311 M * Roey daniel_hozac: you're talking about using openvpn with the server running on a vserver guest?? 1136844329 M * derjohn Roey, why not ssh -X -C ? 1136844343 M * daniel_hozac Roey: yes, i believe so. clients should require no more setup, though. 1136844349 M * Roey derjohn: too slow 1136844357 M * daniel_hozac derjohn: it's -Y now, isn't it? 1136844361 M * Roey oh? 1136844376 M * derjohn Roey, no, slow is the X Protocol not the ssh .... 1136844387 M * derjohn daniel_hozac, hm? /me reads manpage ... 1136844437 M * daniel_hozac derjohn: IIRC the changed the trusted-ness default. 1136844444 M * daniel_hozac +y 1136844453 A * Roey zonks out 1136844455 M * Roey <-- sick 1136844459 M * Roey I'm going home 1136844461 M * Roey see ya there. 1136844478 M * derjohn daniel_hozac, -X Enables X11 forwarding. This can also be specified on a per-host basis in a configuration file. 1136844492 M * daniel_hozac derjohn: two paragraphs below that ;) 1136844493 M * derjohn daniel_hozac, which ssh version are you referring to? 1136844507 M * daniel_hozac 4.2+, as i recall. 1136844524 M * derjohn daniel_hozac, wahts the difference? Both seem to work, or? 1136844549 M * derjohn daniel_hozac, "trused" in which sense? xauth? 1136844581 M * daniel_hozac For this reason, X11 forwarding is subjected to X11 SECURITY extension restrictions by default. Please refer to the ssh -Y 1136844582 M * daniel_hozac option and the ForwardX11Trusted directive in ssh_config(5) for more information. 1136844584 M * derjohn daniel_hozac, I always asked myself why debian ships X with --nolistentcp option turned on 1136844636 M * daniel_hozac X11 SECURITY broke a whole lot of programs. 1136844678 M * daniel_hozac (but, all of this is hear say. i do very little myself, i generally just read the mailing lists :)) 1136844732 M * derjohn daniel_hozac, interesting, soemm days ago i moved my /etc/X11 directory away and wanted to refresh all files ... I didnt manage ist. I think there must be some security foo inside I did miss to create.... 1136844769 M * derjohn daniel_hozac, interesting is that -Y should give you access to the 'original' display ... (i.e. "VNC-like Mode" ?) 1136844818 M * derjohn daniel_hozac, I will try out ... but I wait for Roey to give us the infos about howto-freeNX ;) 1136844850 M * daniel_hozac i've been meaning to look into putting a nx server in a guest myself. 1136844884 M * derjohn cool, if you both success ... dont forget to put the setup-scripts on the wiki :))) 1136845314 Q * bonbons Quit: Leaving 1136845359 Q * lilalinux_ Remote host closed the connection 1136846520 J * tudenbart ~willi@xdsl-213-196-243-136.netcologne.de 1136846750 Q * Johnnie Read error: Connection reset by peer 1136846966 Q * dothebart Ping timeout: 480 seconds 1136847525 M * sizo n8 1136847622 J * bbd_ evaldas@barclay.balt.net 1136847623 Q * bbd Read error: Connection reset by peer 1136848606 Q * emp Ping timeout: 480 seconds 1136848759 P * Hollow SIGTERM 1136848782 J * Hollow ~hollow@home.xnull.de 1136849660 Q * SiD3WiNDR Ping timeout: 480 seconds 1136850067 J * SiD3WiNDR luser@bastard-operator.from-hell.be 1136850185 Q * prae Quit: Pwet 1136850237 N * Bertl_oO Bertl 1136850243 M * Bertl back now ...