1109721614 M * Bertl but let me check the debug output ... 1109721642 M * Bertl no I guess I'm too tired for this right now ... I'm off to bed (will look at it tomorrow) 1109721658 M * Doener hm, after setting the arp entry myself i can reach the guest, but not the host 1109721666 M * Doener sweet dreams Bertl! 1109721759 M * Bertl cya folks ... 1109721766 N * Bertl Bertl_zZ 1109721776 Q * MCLIN Quit: Client Exiting 1109721847 M * Doener seems additions to the local table always go to the host, at least they don't show up on the guest's table 1109722181 M * alexx Doener, have you 5 minutes for me ? 1109722275 M * Doener sure 1109722275 M * alexx great :) 1109722275 M * Doener yep, the routes are added to the host's table if using netlink stuff 1109722275 M * alexx i try to migrate an existing RH7.2 server in a vserver 1109722275 M * alexx but 1109722275 M * Doener as a host or as a guest? 1109722278 M * alexx as a guest 1109722281 M * Doener ok 1109722294 M * alexx rh7.2 will be a guest of a debian 1109722309 M * alexx when i try to start vserver 1109722323 M * alexx i have : vcontext: open(): No such file or directory 1109722336 M * alexx but i don't find the "searched file" 1109722370 M * alexx i have made strace et --debug, but, i haven't more informations 1109722390 M * Doener could you make the --debug output? (e.g. pastebin.com) 1109722408 M * Doener remove sensitive as needed 1109722455 M * alexx alex.ikse.org/debug-vserver 1109722473 M * alexx (too long to copy/paste page by page ;) ) 1109722523 M * Doener /etc/rc.d/rc exists inside the guest? 1109722550 M * alexx if you want color : http://alex.ikse.org/debug-vserver.phps :) 1109722589 M * Doener more important, that one does line-wrapping :) 1109722590 M * alexx /etc/rc.d/rc : yes 1109722677 M * Doener which tools are that? 0.30.204? 1109722702 M * alexx This program is part of util-vserver 0.30.203 1109722711 M * Doener ok 1109722878 M * alexx i build 204 now 1109722907 M * Doener are there other vservers on that box? 1109722914 M * alexx yes 1109722932 M * alexx build with a debian image from the vserver web site 1109722940 M * Doener and that works? 1109722941 M * alexx (lycos-vds image) 1109722949 M * alexx that work fine :) 1109722976 M * alexx do you want a strace of something ? 1109722986 M * Doener they're in the base vserver directory? for example /vservers 1109723009 M * alexx yes 1109723016 M * alexx with debian specification ;) 1109723024 M * alexx /var/lib/vservers 1109723038 M * alexx but, i have make a link /vservers for simplify 1109723047 M * Doener hm... try to strace -fF that looong line above the error message 1109723070 M * Doener but i kinda doubt that strace will be able to follow the context change 1109723082 Q * ciphernaut Quit: 1109723090 M * alexx want a vstrace ? ;) 1109723102 M * Doener there's such a tool? 1109723115 A * Doener never tried stracing the startup stuff 1109723120 M * alexx yes, in the future ;) 1109723308 M * alexx build of 204 finish, but, same error 1109723437 M * alexx http://alex.ikse.org/strace.vserver.phps 1109723456 M * Doener yep, expected that 1109723696 J * c1pher0 ~a@61.88.18.130 1109723744 N * c1pher0 ciphernaut 1109723975 M * alexx Doener, i've do that with success under linux2.4 with last stable 1.2.x 1109724003 M * Doener proc is unhidden? 1109724009 M * alexx yes 1109724113 M * Doener is /dev/null available in the vserver? 1109724188 M * Doener hm, doesn't matter 1109724266 M * alexx ho ......... Doener 1109724278 M * alexx you are really a great guy ! 1109724288 M * Doener hu? it was /dev/null? 1109724293 M * alexx yes :) 1109724318 M * alexx i must kick my own ass for this forgot :( 1109724325 M * Doener strange... it is the only thing i could find in the source that could cause the error, but it works just fine without it here... 1109724334 M * alexx what is the minimim /dev file i must have ? 1109724372 M * Doener full null ptmx pts random tty urandom zero 1109724557 M * alexx with dev ... it work better ... ;) 1109724587 M * Doener ok, great 1109724753 M * Doener back to the routing stuff then :) 1109724782 M * alexx ok, if you want a dumb for test ... i'm you man :) 1109724788 M * alexx +r 1109727710 J * duckx ~Duck@dyn-83-157-149-170.ppp.tiscali.fr 1109730061 J * zimdog ~zimdog@c-67-164-190-201.client.comcast.net 1109730108 M * zimdog Hello everyone 1109730224 M * Doener welcome zimdog 1109730268 Q * duckx Read error: Operation timed out 1109730280 M * zimdog How does one go about creating a vserver image? 1109730318 M * Doener there are no 'images' like in vmware (just to avoid confusion ;) 1109730357 M * zimdog Well I am confused :) 1109730366 M * Doener heh 1109730403 M * Doener linux-vserver is kind of an enhanced chroot, so just put the vserver's file into the vserver's root directory 1109730424 M * Doener the alpha tools also contain a nice build method for some distros 1109730436 M * Doener http://linux-vserver.org/alpha+util-vserver 1109730473 M * zimdog I was looking at the zipped up images available on the site. 1109730505 M * zimdog Can I copy a kernel over and boot from it and then run kickstart or something? 1109730550 M * Doener linux-vserver is not about full (or even partial) system emulation, but about sharing ressources 1109730589 M * Doener there's just one kernel on a linux-vserver box, the vservers run in a so-called security context that isolates them from the host and the other vservers 1109730606 M * zimdog Sorry I am really new to this. 1109730632 M * Doener if you want to have a look at the basic concepts (and find out if it is the right thing for you) we got a nice paper... http://linux-vserver.org/Linux-VServer-Paper 1109730750 M * zimdog I am sure this is what I want. I havbe it installed and running the debian and redhat9 image from the site. I thought these images had a /boot directory etc. and was wondering how they were created. Guess I need to do some more reading. 1109730751 M * Doener Bertl_zZ: ok, i managed to make the routing stuff behave :) the single netlink socket was never tagged. For now, i've simply hacked skb tagging into netlink's rcv_skb 1109730785 M * Doener probably someone just didn't remove those directories, or the distro's aren't happy without them 1109730850 M * zimdog Ok thanks for the help 1109730889 M * Doener a vserver gets (by default) started by just invoking the rc script with runlevel 3 1109730914 M * Doener you can change that to use init or whatever you want 1109730944 M * Doener just 'real' booting using a kernel isn't possible 1109730971 M * zimdog Soa can the vserver context run without anything in it's root directory? 1109731015 M * Doener the context: yes, if you also want to chroot: of course not ;) 1109731068 M * Doener the "vserver xxx start/stop/enter" is just some clever program that combines a lot of different things to create what we call a vps or vserver 1109731105 M * Doener the context is just the part that isolates the vserver on process level 1109731217 M * zimdog So could I copy a a kickstart disk into the directory of the vserver and then enter the vserver and run it to install the packages I want availble for the vserver? 1109731256 M * Doener what's a kickstart disk? reminds me of the amiga 500 days ;) 1109731336 M * zimdog Basically it installs and configures a system by reading a file that tells what packages to install etc. 1109731351 M * zimdog Redhat,centos Mandrake use them 1109731355 M * Doener Bertl_zZ: http://doener.homeip.net/doener/vserver/diff-2.6.11-rc5-vs1.9.4.9-ng9.1-hack_netlink.diff 1109731360 M * daniel_hozac util-vserver's rpm build method should work for that. 1109731380 M * zimdog Ok i will look for that then 1109731388 M * Doener ah, my saviour, someone who knows rpm stuff :) 1109731447 M * zimdog Is util-vserver part of the tools that are installed by default? 1109731462 M * Doener util-vserver _is_ the tools ;) 1109731579 M * zimdog Ok thanks for the help. I need to get more familiar with this. I willl just hang out and watch and learn. 1109731591 M * Doener you're welcome 1109732365 M * zimdog One more question. When I installed thsi I downloaded the kernel source. Applied the patch from 13thfloor.at. When I did the make config I did not see some of the options mentioned like File systems/Persistient ContextID and this I believe Quota suboption/VFS vo quota format support. Is there more than just the linux-vserver-1.2.10.tar patch I need to install? 1109733524 M * Doener no idea where that stuff is on 2.4 kernels, i'm doing 2.6 stuff only, sorry 1109734197 M * zimdog How functional is the 2.6 stuff? 1109734961 M * mugwump on a scale from perl 6 to xterm, it's about gcc 1109735067 J * locksy ~locksy@mrtg.sisgroup.com.au 1109735292 M * ciphernaut using vserver v. 0.30.204, what is the best way to hook in post install scripts? (As opposed to --post-install-script in the stable version) 1109735320 M * ciphernaut s/install/build 1109735328 M * locksy How do I bind mount a root vserver directory (/var/run/mysqld) into a vserver. Kernel: 2.6.10-vs1.9.4 VSAPI:0x10025 util:0.30.196 1109736017 M * ciphernaut locksy: what are you trying to do? 1109736085 M * ciphernaut i run one copy of mysql server on the host system, the other systems can just connect to it over tcp/ip 1109736130 Q * aba Ping timeout: 480 seconds 1109736294 M * mugwump locksy: or, use mount --bind /var/run/mysqld /vserver/foo/var/run/mysqld, but be careful of what happens when vservers remove that file 1109736511 M * locksy where should I put the 'mount --bind /var/run/mysqld /vserver/foo/var/run/mysqld'? 1109736549 M * mugwump if you're using newer util-vserver, the correct place is somewhere like /etc/vservers/foo/scripts/pre-start 1109736558 M * mugwump or /etc/vservers/.defaults/scripts/pre-start 1109736578 M * locksy 0.30.196 new enough? 1109736581 M * mugwump yes 1109736642 M * locksy Do I have to turn on/off some flag for namespaces? 1109736745 M * mugwump yes, put "namespace" in the `flags' file 1109736752 M * mugwump I use a lot of flags; 1109736758 M * mugwump nproc 1109736758 M * mugwump ulimit 1109736758 M * mugwump namespace 1109736758 M * mugwump sched_prio 1109736758 M * mugwump virt_mem 1109736760 M * mugwump virt_uptime 1109736763 M * mugwump virt_cpu 1109736765 M * mugwump hide_mount 1109736768 M * mugwump hide_netif 1109737220 J * aba ~aba@sol.turmzimmer.net 1109737599 J * AzAr ~AzAr@S0106000103c53368.ed.shawcable.net 1109737669 M * AzAr looking for realy bot for multiserver any help ? 1109738078 Q * AzAr Quit: cya latter 1109739216 Q * nox Ping timeout: 480 seconds 1109741024 Q * zimdog Quit: Leaving 1109741272 J * chairuou ~chairuou@210.245.74.170 1109742272 M * micah ciphernaut: I actually found it better to create 127.0.0.x interfaces in all my vservers and access mysql over that network 1109742302 M * micah ciphernaut: because if you mount --rbind your /var/run/mysql into other vservers, your sockets will get removed when you start vservers 1109743552 M * ciphernaut I configured mysql to listen on the hosts' ip address ... I'm not sure why I would want to mount /var/run/mysql at all .... is that for unix socket based comms? 1109743676 M * ciphernaut i imagine that my setup wil be more complex to firewall later on 1109744894 Q * ciphernaut Ping timeout: 480 seconds 1109748676 Q * _are_ Quit: Disconnecting 1109749577 J * ciphernaut ~a@61.88.18.130 1109749896 J * erwan_ho ~erwan@lns-vlq-39f-81-56-133-136.adsl.proxad.net 1109749981 Q * logger Ping timeout: 480 seconds 1109750176 Q * erwan_ho Remote host closed the connection 1109750425 Q * lilo Read error: Operation timed out 1109750688 J * lilo ~lilo@lilo.usercloak.oftc.net 1109751215 J * _are_ ~are@gateway-dsl.lihas.de 1109751365 J * logger ~rs@vds.pas-mal.com 1109751425 N * Doener Doener_zZz 1109751592 Q * lilo Read error: Operation timed out 1109752953 J * prae ~prae@ezoffice.mandrakesoft.com 1109754702 N * Bertl_zZ Bertl 1109754714 M * Bertl morning folks! 1109754739 M * alexx morning Bertl :) 1109754808 M * prae Hi alexx 1109754811 M * prae Hi Bertl 1109754830 M * Bertl greetings alexx, prae! 1109755216 M * erwan_taf lo Bertl 1109755522 M * Bertl morning erwan_taf! 1109755729 M * alexx morning erwan_taf 1109755783 M * erwan_taf lo alexx 1109756668 M * prae hi erwan_taf :) 1109756683 M * erwan_taf prae 1109757194 J * rs ~rs@194.98.28.2 1109757239 M * rs hi there 1109757245 M * Bertl morning rs! 1109757281 M * rs how are you bertl? 1109757305 M * Bertl it's okay I guess (too early to tell ;) 1109757311 M * Bertl and you? 1109757353 M * rs you just waked up? :) 1109757360 M * Bertl unfortunately not ;) 1109757390 M * rs ok, didn't sleep yet ? :) 1109757408 M * Bertl I did sleep, but I had to get up early today ... 1109757421 M * Bertl (and I'm still not fully awake yet ;) 1109757434 M * rs I can understand 1109757489 M * Bertl so what are your plans regarding testing (in the near future?) 1109757567 M * Bertl (if there are plans at all) 1109757568 M * rs I will finish the IP migration today, then I will setup a new machine for you (to replace the controle, it won't be accessible anymore), then I can work on ngnet 1109757605 M * rs controle/controller 1109757637 M * rs did you manage to fix the hard scheduler bug ? 1109757833 M * Bertl well, I do not know what the bug is yet ... 1109757869 M * Bertl if you tell me, I'll fix it immediately ;) 1109757893 M * rs though you found the source 1109757898 M * rs of the issue 1109757918 M * Bertl no, I just see strange behaviour on 'some' machines ... 1109758620 M * matti Bertl: :0 1109758621 M * matti ;] 1109758637 M * Bertl morning matti! 1109758875 M * matti :> 1109758883 M * matti What's up Bertl? 1109759067 M * Bertl not much ... 1109759710 Q * _are_ Quit: Disconnecting 1109759827 J * nox ~nox@80.171.3.221 1109759867 M * Bertl welcome nox! 1109759898 M * Bertl okay, off for lunch, back later ... 1109759903 M * erwan_taf ++ 1109759907 N * Bertl Bertl_oO 1109761383 Q * chairuou Ping timeout: 480 seconds 1109761526 J * atsab ~as@lotes.vtu.lt 1109761545 M * atsab anyone tried vserver+plesk? 1109761554 M * atsab any success? 1109761842 N * Bertl_oO Bertl 1109761858 M * Bertl welcome atsab! 1109761871 M * Bertl did you try it? 1109761886 M * atsab no, just about-to-try ;-) 1109761903 M * atsab is there any search engine on irc logs? 1109761912 M * Bertl google? 1109761926 M * atsab 1 mention about subj. only 1109761954 M * Bertl probably not that interesting ;) 1109762070 J * lilo ~lilo@lilo.usercloak.oftc.net 1109762117 M * Bertl wb lilo! 1109762128 M * atsab but: lesk Server Administrator 1109762129 M * atsab (PSA) 5.0 hosting automation platform with the Vserver open-source 1109762129 M * atsab VPS technology 1109762136 M * atsab it was in 2002 1109762141 M * atsab so its possible? 1109762161 M * Bertl atsab: if you encounter issues which are general issues (i.e. not plesk specific) we'll gladly fix them ... 1109762182 M * atsab huh, got it 1109763114 M * prae its exists a patch with vserver+grsec for 2.6 ? (I don't think so... the latest version is for 2.4.29, and it's my patch) 1109763174 M * Bertl I remember somebody doing such a patch ... 1109763202 M * prae ah ? 1109763236 M * prae You remember where you saw it? 1109763240 M * prae into website, ML ? 1109763259 M * TheSeer Bertl: do we have a a patch for 2.6.11 yet? 1109763266 M * Bertl prae: check the irc log ... 1109763271 M * prae ok 1109763281 M * prae all IRC logs ?!! 1109763281 M * Bertl TheSeer: I hope so ... according to Linus rc5 should work fine ... 1109763291 M * Bertl TheSeer: will know in a few minutes ;) 1109763296 M * TheSeer :) 1109763315 M * TheSeer i'm pretty eager to update since my core server is running 2.6.10 atm 1109763334 M * TheSeer and it died a second time for no aparent reason last weekend 1109763349 M * Bertl do you use hard cpu scheduling? 1109763411 M * TheSeer ouhm.. even i might a fool out of my self by asking.. but how do i tell? *g* 1109763428 M * Bertl if you don't know, you're not using it ;) 1109763449 M * Bertl but to answer your question: 1109763474 M * Bertl it's a flag and a kernel option, the flag is called sched_hard and the kernel option *HARD_CPU* 1109763573 M * prae " it seems that nobody adapted any grsec patches for 2.6.10 or .11 yet " :) 1109763629 M * prae no grsec+vserver patch found :| 1109763631 M * prae :'( 1109763639 M * prae so ... let's rock ! 1109763652 M * prae I will do it :p 1109763728 M * Bertl don't let me stop you ;) 1109763754 M * prae ;-)) 1109763983 M * Bertl TheSeer: http://vserver.13thfloor.at/Experimental/patch-2.6.11-vs1.9.5-rc1.diff 1109764063 M * TheSeer thanx :) 1109764070 M * Bertl my pleasure! 1109764096 Q * atsab Ping timeout: 480 seconds 1109764782 M * TheSeer Bertl: i recall having a mem-split patch applied at some point... 1109764803 M * TheSeer Bertl: do i still need it or, to be exact, is it recommended? (the box has 2 GB Ram..) 1109764820 M * Bertl it's included in linux-vserver now, so just select 2/2 as split 1109764831 M * TheSeer k.. nice :-) 1109764833 M * Bertl or maybe 1/3 even ... 1109767895 J * FEN_HIN ~JFOC@latarius.tkdgroup.com 1109767906 M * Bertl welcome FEN_HIN! 1109767912 M * FEN_HIN thx Bertl 1109767929 M * FEN_HIN have you ever experience with linux-vserver? 1109767945 M * Bertl experienced what? 1109767951 M * eyck nope, Bertl is a NEWB around here 1109767955 M * FEN_HIN using it 1109767958 M * FEN_HIN oh ic 1109767962 M * TheSeer *lol* 1109767970 M * eyck don't talk to him, he'll just mess with your head 1109768014 M * Bertl right ;) 1109768027 M * FEN_HIN i'm just want to know from every experience vserver user, what best between linux-vserver and uml 1109768031 M * FEN_HIN :D 1109768040 M * Bertl but eyck knows a lot! just has not much time for stuff ... 1109768057 M * FEN_HIN ohic 1109768061 Q * FEN_HIN Quit: 1109768064 M * eyck that's not true!, just the other day I had literally 15 minutes of time! 1109768078 M * eyck FREE time 1109768101 A * Bertl is now checking eycks time in the irc logs ... 1109768103 M * eyck I couldn't sleep and woke up too early.... :( 1109768113 J * FEN_HIN jfoc@admin.teamshell.net 1109768118 M * Bertl wb FEN_HIN! 1109768124 M * FEN_HIN sorry disconnected 1109768135 M * FEN_HIN Thank's 1109768135 M * eyck no problem, go ahead, ask your question. 1109768493 M * FEN_HIN Thank's 1109768496 M * FEN_HIN i'm just want to know from every experience vserver user, what best between linux-vserver and uml 1109768532 M * Bertl linux-vserver and uml are pretty orthogonal (different, unrelated) things 1109768538 M * eyck I, from my personal experience can honestly tell you, that linux-vserver is the bestest. 1109768554 M * eyck see, I told you Bertl will try to mess with your head. 1109768557 M * Bertl you can use UML inside linux-vserver guests or put linux-vserver kernel in UML ... 1109768568 M * FEN_HIN oh i see 1109768600 M * eyck it's just uninteligible dribble he's trying to force upon you, my answer is clear, precise and generaly the best. 1109768615 M * TheSeer darn.. the box doesn't come back up.. 1109768618 M * TheSeer i HATE that.. :// 1109768631 M * Bertl but of course, if eyck says linux-vserver is the 'bestest', I can't hardly disagree ... 1109768639 M * eyck damn right. 1109768668 M * prae fusion grsec+vserver : done 1109768680 M * FEN_HIN hehehe 1109768685 M * Bertl prae: was it a cold or a hot one? 1109768705 M * FEN_HIN i'm try to make private server to my debian box 1109768739 M * eyck hmm, or was it 'damn straight home-boy'? 1109768750 M * eyck I can't recall what language I was trying to speak... 1109768751 M * eyck write.. 1109768759 M * FEN_HIN Laughing Out Loud 1109768760 M * prae Bertl: I test on my own servers 1109768796 M * prae Bertl: I says if it's fusion is ok or not just after that :) 1109768806 M * prae s/it's/this/ 1109768863 M * eyck oh well, stop with colours please. 1109768881 M * Bertl FEN_HIN: you like Color? 1109768917 M * prae =) 1109768963 M * FEN_HIN yes i like it 1109768975 M * FEN_HIN make my life more colourfull 1109769025 M * prae pouarf :) 1109769232 M * SiD3WiNDR :D 1109769271 M * prae funnnyy ! 1109769312 M * FEN_HIN heuheuheuhe 1109769494 Q * erwan_taf Quit: Leaving 1109769536 J * erwan_taf ~erwan@81.80.43.68 1109769536 P * erwan_taf 1109769593 M * FEN_HIN silent night herE? 1109769663 M * prae silent for hard working 1109769664 M * prae :) 1109769694 M * prae s/ing$/s/ 1109769755 M * FEN_HIN prae: experience installing vserver in debian ? 1109769821 M * FEN_HIN i'm using kernel 2.4.27 and 2.4.29 1109770140 M * prae yes 1109770145 M * prae 2.4.x and 2.6 1109770183 M * prae grsecurity/gracl_shm.c: In function `gr_handle_shmat': 1109770183 M * prae grsecurity/gracl_shm.c:21: warning: implicit declaration of function `vx_rmap_pid' 1109770184 M * prae hmmm ... 1109770198 A * prae note this 1109770204 M * FEN_HIN hm 1109770242 M * prae FEN_HIN: nothing, it's juste for me :) 1109770245 M * prae -e 1109770253 M * FEN_HIN :) 1109770303 M * TheSeer Bertl: the 1.9.5 patch doesn't show in the kernel version string, or does it? 1109770397 M * Bertl well, it's supposed to, but it seems it does not ... 1109770409 A * Bertl is off bashing on that cleanup script ... 1109770430 M * TheSeer uname -r 2.6.11 1109770489 M * FEN_HIN Bertl: you know where i can download vserveradmin ? i searching from linux-vserver but couldn't find 1109770525 M * Bertl what is vserveradmin? 1109770536 M * SiD3WiNDR FEN_HIN: night? :p 1109770553 M * SiD3WiNDR hmm 1109770561 M * SiD3WiNDR is it possible to make a UML with vservers inside? :P 1109770578 M * Bertl read up, I said that a few lines before ... 1109770612 M * SiD3WiNDR hah 1109770620 M * SiD3WiNDR indeed, right out of my scrollbak 1109770622 M * SiD3WiNDR +c 1109770660 M * SiD3WiNDR it's either vmware-on-windows with 2.6+vs+ng or uml with 2.6+vs+ng ... what would be best? :) 1109770682 M * Bertl UML has lower overhead than vmware ... 1109770697 M * Bertl linux has higher performance than win* 1109770727 M * Bertl but I guess you want a business solution, so what about MS-Vserver(TM,R) 1109771034 M * eyck hmm, I think I saw linux-on-vmware performing better then uml 1109771074 M * eyck but AFAIK only vserver is viable solution for performance-sensitive virtualisation solutions 1109771100 M * eyck this will change when xen+vanderpool become stable and available 1109771248 M * SiD3WiNDR Bertl: no, I want to test 2.6 vserver with ngnet :) 1109771265 M * SiD3WiNDR and I just dont have any spare hardware for yet another box, so it should be virtualized ;) 1109771279 M * FEN_HIN SiD3WiNDR: sorry, i was talking with my friend. yes it's night at my country 1109771282 M * Bertl well, then use QEMU for that? 1109771289 M * SiD3WiNDR hmm, I don' tknow qemu :) 1109771297 A * SiD3WiNDR googles 1109771308 M * SiD3WiNDR FEN_HIN: where are you? it's almost 3pm here :) 1109771514 J * FEN_HIN__ jfoc@admin.teamshell.net 1109771516 Q * FEN_HIN Remote host closed the connection 1109771525 M * FEN_HIN__ sorry disconnect 1109771530 N * FEN_HIN__ FEN_HIN 1109771541 M * FEN_HIN SiD3WiNDR: i'm from indonesia 1109771584 M * FEN_HIN hmmm strange always disconnect if i've been idle 1109772282 M * TheSeer sounds like a broken NAT in between ;) 1109772302 M * FEN_HIN oh.. 1109772318 M * FEN_HIN but i dont have this problem at another irc 1109772983 M * TheSeer dunno.. was just a wild guess.. 1109773059 M * FEN_HIN hehehe 1109773192 Q * monrad Quit: Leaving 1109773219 Q * Bertl Read error: Operation timed out 1109773362 J * Bertl ~herbert@janus.mc.tuwien.ac.at 1109773566 M * FEN_HIN wb Bertl 1109773627 J * Hollow ~Hollow@home.xnull.de 1109775197 Q * FEN_HIN Remote host closed the connection 1109775283 J * FEN_HIN jfoc@admin.teamshell.net 1109775302 M * Hollow Bertl: the 1.9.5_rc1 patch doesn't change EXTRAVERSION 1109775740 M * Bertl ah, back now ... 1109775761 M * Bertl (regular router outage :/) 1109775777 M * Bertl thanks Hollow! 1109776045 M * Loki|muh why dont you schedule it at 6a.m. or so? 1109776112 M * Bertl well, it's not up to me to 'schedule' it ;) 1109776132 M * Bertl (and I do not think it's intentional either ;) 1109776195 J * Pazzo ~thomas@adsl136-175.aknet.it 1109776198 M * Bertl okay, 2.6.11-vs1.9.5-rc1 is updated (i.e. changes the version) 1109776202 M * Bertl welcome Pazzo! 1109776225 M * Pazzo hi pazzo! 1109776234 M * Bertl hmm ... *G* 1109776236 M * Pazzo ehm 1109776243 M * Pazzo hi bertl :) 1109776306 M * Pazzo do you need some testing for 1.9.5-rc1 or will 1.9.5 be release somewhere around midnight? 1.9.4.x seems pretty stable :) 1109776330 M * Bertl testing would be nice, but I consider it pretty stable too ... 1109776348 M * Bertl (unless you use the hard cpu scheduler, which probably has a bug somewhere) 1109776410 M * Pazzo nope - I would really like to dig deeper into all this ressource limiting / cpu scheduling stuff - but can't find the time to do so atm 1109776429 N * Doener_zZz Doener 1109776454 M * Doener morning! 1109776472 M * Bertl morning Doener! 1109776491 M * Bertl back in a minute, have to logout to remove some modules ... 1109776502 N * Bertl Bertl_oO 1109776536 M * Pazzo Bertl: I have to remotely reinstall a friends server tonight - it will be a great playground for testing rc1 as these guys would really like to run vservers on a 1 TB crypto device on software raid5 with 1gb flat uplink *gg* no idea what it will be used for :-) 1109776574 J * Tauras ~gregory@195.158.80.126 1109776581 T * Doener http://linux-vserver.org/ | latest stable 1.2.10, devel 1.9.5-rc1, ng9.1 -- 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 ;) 1109776595 M * daniel_hozac forgive my ignorance, but why is lowmem better than highmem? (i.e. why would i use a different memory split rather than enabling highmem?) 1109776637 M * Pazzo hi doener! 1109776645 M * Doener hi Pazzo 1109776666 M * Tauras hello everyone 1109776673 M * Doener welcome Tauras 1109776757 M * Tauras where can I found how to configure vserver for 1.9.4 utils? I mean where to put values frm .conf file in config directory? 1109776770 M * FEN_HIN nice wiki 1109776773 M * FEN_HIN i love wiki 1109776869 M * Doener Tauras: there's the flower page, which describes the new configuration. Either use the file in your util-vserver's doc/ or http://www.tu-chemnitz.de/~ensc/util-vserver/doc/conf/configuration.html 1109776913 M * Tauras thank you 1109776923 M * Doener Tauras: it's easier to use the tools' skeleton build method to create a basic configuration and adjust that to your needs than creating one from scratch 1109776994 M * Doener take a look at "vServer Creation" on http://linux-vserver.org/alpha+util-vserver if you don't know about the build stuff yet 1109777027 M * Tauras yes, followed this page to create new debian vserver 1109777033 M * rs re 1109777045 M * Pazzo hi rs! 1109777049 M * rs hey Pazzo! 1109777060 J * monrad ~monrad@213083190130.sonofon.dk 1109777064 M * Doener ok, then you should now have everything to start playing around with the new config scheme :) 1109777169 M * Doener prae: IIRC DaPhreak has been working on vserver+grsec patches 1109777301 M * prae sure ? 1109777303 M * prae for 2.6 ? 1109777318 M * Doener yep 1109777321 M * prae ?!! 1109777331 M * prae where ? 1109777334 M * prae :'| 1109777396 M * Doener i don't know where he worked on them ;) and i also don't know if he released anything yet 1109777483 M * FEN_HIN Doener: you know where's i can find documentation to create debian vserver using kernel 2.4.x ? and vserveradmin i can downloaded 1109777485 M * FEN_HIN thx :D 1109777589 M * Doener i never used that vserveradmin stuff... just debootstrap a debian base-system into the vserver's root directory and create a configuration... 1109777664 M * FEN_HIN oh i see 1109777689 M * FEN_HIN so after i build vserver bootimage, i can using vserver stuff ? 1109777696 M * FEN_HIN like administrating vserver? 1109777779 M * Doener there's no "bootimage" for vserver's. they're started by running their rc script at runlevel 3 (by default) 1109777819 M * FEN_HIN Oh I See 1109777820 M * Doener a nice description of the old configuration scheme is here: http://linux-vserver.org/linux-vserver_administrators_guide 1109777824 M * Doener (section 2) 1109777855 M * FEN_HIN is it still working with new vserver? 1109777860 M * Doener AFAIK you could also just use the alpha tools with your 2.4 kernel and use their build methods 1109777891 M * Doener with 2.6 kernels you should use the alpha tools, otherwise you'll miss all those new features ;) 1109777905 M * FEN_HIN :) 1109778027 M * FEN_HIN vserver-0.32src is newest one for 2.4.x 1109778029 M * FEN_HIN ?¿?¿?¿?¿?¿?¿?¿ 1109778085 M * Loki|muh FEN_HIN: this sucks 1109778115 M * FEN_HIN Loki|muh: suck? why ? have any good reason for that? 1109778137 M * Loki|muh it is irritating 1109778169 M * FEN_HIN hmm ? 1109778217 M * Doener because even having more than one (or maybe two) questionmarks at the end of a sentence already sucks... 1109778270 M * Doener and you got 14 without a sentence... (maybe they're even colored?) 1109778290 M * FEN_HIN oh i see, sorry for my poor english 1109778303 A * Doener .oO( color-stripping can save your eyes! ) 1109778310 M * FEN_HIN because i'm indonesian, and not fluently in english 1109778357 M * FEN_HIN sorry Doener, Loki|muh, or anybody if you feel it ... 1109778397 M * Loki|muh yeah, thanks ;) 1109778416 M * Loki|muh using colors in normal chat is not considered as nice ;) 1109778419 M * FEN_HIN thank you for your good kindness 1109778428 M * Loki|muh poor english doesnt matter ;) 1109778440 M * Loki|muh i think the most of us arent native speakers ;) 1109778442 M * Doener anyway, for that question: vserver-032src is from Jack (the one who started the vserver stuff back then). I have no idea about it because we're pretty much using util-vserver only these days. util-vserver is a replacement for the vserver-tools... 1109778456 M * eyck yup, native english speakers can't code. 1109778478 M * FEN_HIN Loki|muh: oh i see 1109778493 M * FEN_HIN Doener: thank you for your answer 1109778499 M * Loki|muh eyck: lol 1109778511 M * FEN_HIN hahaha 1109778523 M * FEN_HIN eyck: little agree with you ... i think :p 1109778600 M * eyck well known fact. 1109778682 M * FEN_HIN Doener: is http://www-user.tu-chemnitz.de/~ensc/util-vserver/doc/virtual-servers.pdf is tutorial for creating vserver? 1109778763 M * Doener no, those are slides from a speech that enrico (creator of util-vserver) held 1109778788 M * FEN_HIN oh 1109778815 M * FEN_HIN sorry i just downloading everything after that i will read all of it 1109778883 M * Doener if you didn't setup any vservers yet, you may want to (try to) uninstall the vserver-032src thing and vserveradmin or whatever you installed and try out the alpha tools 1109778895 Q * Dus10 Quit: [BX] Been around the world and found that only stupid people are breeding. 1109778905 M * Doener http://linux-vserver.org/alpha+util-vserver -- here you can find some documentation on building a vserver using the alpha tools 1109778926 M * Doener they also contain a build method for debian vservers 1109778928 M * FEN_HIN yes, not yet make any configuration 1109778959 M * Doener (the [alpha branch] thing in the heading is a link to the download location for the alpha tools ;) 1109779014 M * FEN_HIN thx i will see it 1109779458 J * atsab ~as@lotes.vtu.lt 1109779617 M * Doener Bertl_oO: http://doener.homeip.net/doener/vserver/diff-2.6.11-rc5-vs1.9.4.9-ng9.1-various.diff 1109779684 P * Pazzo 1109780094 M * prae Bertl_oO: so, kernel compile perfectly 1109780108 M * prae Bertl_oO: and start perfectl too 1109780475 M * Doener interesting... no changelog for 2.6.11 on kernel.org 1109780490 M * FEN_HIN wow 1109780523 M * prae Doener: no kidding ? ô_O 1109780563 Q * FEN_HIN Remote host closed the connection 1109780620 M * Doener no link on the kernel table, but there's one on the ftp server 1109781061 M * prae http://team.lea-linux.org/bgigon/vserver/linux-2.6.10-grsecurity-2.1.1-2.6.10-as2-200501242254-Vserver-1.9.4.patch.gz 1109781068 M * prae http://team.lea-linux.org/bgigon/vserver/linux-2.6.10-grsecurity-2.1.1-2.6.10-as2-200501242254-Vserver-1.9.4.patch.md5sum 1109781112 M * prae (If Bertl_oO says to another personn to see irc log ;) 1109781299 Q * jsambrook Quit: Download Gaim: http://gaim.sourceforge.net/ 1109781396 J * jsambrook ~jsambrook@aelfric.plus.com 1109781767 P * halorgium hrmm 1109782161 M * Doener wow, if all of Bertl's minutes are that long, he'll probably live forever :) 1109782300 M * daniel_hozac hehe 1109782485 A * Doener has to admin that some of his 'seconds' were even longer ;) 1109782555 M * Doener s/admin/admit/ 1109782788 J * mhepp ~mhepp@r30s12p13.home.nbox.cz 1109784350 M * Tauras is it possible to have lo interface inside vserver? 1109784696 M * eyck you ain't got one? 1109785492 M * Tauras not in vserver 1109785550 M * Tauras and what about routes? ip route ls shows host server route and src ip 1109786591 M * prae Tauras: loopback iface ? yes, it's possible 1109786631 M * lilo (thanks Bertl) 1109786638 Q * lilo Quit: bbiab 1109786680 M * prae Tauras: I use it into my vservers to talk between them. 1109786726 J * hillct ~hillct@client200-5.dsl.intrex.net 1109786751 M * daniel_hozac prae: that's not really necessary. 1109786779 M * prae I know 1109786781 Q * rs Quit: leaving 1109786786 M * prae but I use this method :) 1109786883 J * lilo ~lilo@lilo.usercloak.oftc.net 1109786905 M * IrishKitty hi lilo 1109786944 M * Tauras so how can I configure loopback iface? 1109787128 M * Doener Tauras: you don't have to. Anything that is coming from and going to the box itself is routed through loopback automatically no matter what (of course you can break that with some iptables rules, but that's not what you want ;) 1109787255 M * Tauras so the main tool with iptables is to use -d vIP ? 1109787270 M * Doener hm? to do what? 1109787305 M * Tauras I mean firewall should be configured on host (?) 1109787316 M * Doener yep 1109787326 M * prae so, bye bye everybody :) 1109787330 M * prae see'ya tomorrow 1109787331 M * Doener bye prea 1109787338 M * Tauras bye prae 1109787379 M * Tauras so input IFACE is actually only eth0 ? 1109787385 M * Tauras what about aliases? 1109787412 M * Doener if it's coming from the outside, yes 1109787423 Q * prae Quit: Client exiting 1109787433 M * Tauras let say I want to reject access on port 25 for host (eth0) but allow for vserver (eth0:0) 1109787463 M * Doener eth0 is the same device as eth0:0 1109787483 M * Doener you got to do that based on the ip address 1109787518 M * Tauras ok :-) that was " main tool with iptables is to use -d vIP" 1109787543 M * hillct could someone point me to documentation on vserver --unify/deunify ? 1109787557 M * Tauras ie recognise packages by destiantion IP address 1109787613 M * Tauras btw, Doener, your help highly appreciated, I managed finally to run my first vserver :-) 1109787639 M * Doener you're welcome! 1109787738 M * DaCa say I have a partition mounted r/w in the root server and I also want to mount it r/o somewhere in a vserver, whats the best way to do that? 1109787761 M * Doener probably using BME 1109787797 M * Tauras BME? 1109787803 M * DaCa if I mount it regularly it seems to default to a bind mount, but if I add -o ro it says its already mounted 1109787844 M * Doener Bind Mount Extensions, they allow for example to make bind mounts read-only although the original mount is r/w 1109787844 M * DaCa BME? 1109787870 M * DaCa its a kernel patch? 1109787877 M * Doener yep, from Bertl 1109787881 M * DaCa i am using 2.4 btw 1109787918 M * Doener bme is 2.6-only AFAIK 1109787923 M * Doener oh, no... 1109787932 M * Doener http://www.13thfloor.at/patches/ 1109787936 M * DaCa I was afraid so, thats why I mentioned it 1109787971 M * Doener or here: http://vserver.13thfloor.at/Experimental/BME/ 1109787985 M * Doener dunno if they apply to the latest 2.4 kernels 1109788061 M * DaCa I see one for 2.4.27, I'll check if it applies to .29 1109789269 M * Tauras why there are some modules [permanent] in lsmod output? 1109789293 M * Tauras why they cannot be removed? 1109789341 M * Tauras eg. sis5513 13576 0 [permanent] 1109790091 Q * jsambrook Quit: Download Gaim: http://gaim.sourceforge.net/ 1109790109 J * jsambrook ~jsambrook@aelfric.plus.com 1109790113 P * jsambrook 1109790190 M * Tauras heh I found... locking is still in kernel development for ide drivers 1109790251 M * hillct I have yet another question abbout immutable link invert 1109790268 M * hillct in attempting to create a reference vserver 1109790283 M * atsab Tauras: lietuvis kiba? 1109790285 M * hillct that I can then use to create new vservers from... 1109790316 M * hillct I find that once imunlink has been set, I can no longer create new hard links to that inode 1109790358 M * hillct a such, I have to remove that attribute for each file for each new hard link (neew vserver) I create at the time of creation 1109790368 M * hillct then go back through and reset th attribute 1109790409 M * hillct It seems I must be missing something significant here, as this approach seems frought with problems 1109790595 M * Tauras atsab: ziurek, tu jau ne vienas ;-) 1109790623 J * erwan_ho ~erwan@lns-vlq-39f-81-56-133-136.adsl.proxad.net 1109791549 J * prae ~prae@sherpadown.net 1109792013 M * micah oh wow, 2.6.11 is out! 1109792034 M * micah I knew the day after I patch it would come out :) 1109792056 M * hillct the kernel or vserver patches? or both? 1109792058 M * micah dont suppose anyone has tried the vserver patch against this kernel yet have they? 1109792076 M * micah hillct: i just installed 2.6.11-rc5 yesterday 1109792095 M * hillct ah 1109792107 M * hillct er, I remember you mentioned that yesterday 1109792137 M * hillct bbl 1109792141 P * hillct 1109792571 J * sith sith@aaronp.com 1109792721 M * sith hey guys, I'm just trolling through email lists, and the wiki, but does anyone know what the answer is for someone trying to get vserver working with the newer glibc non-exec versions? 1109792769 M * sith i've tried exec=off and exec32=off on boot, but that doesn't seem to do it. 1109792824 M * sith is there an answer other than change glibc? (ie, some way to disable non-exec stack, etc) 1109792897 M * Tauras bye everyone 1109792907 P * Tauras 1109794240 N * Bertl_oO Bertl 1109794265 M * Bertl evening folks! 1109794269 M * eyck evening 1109794274 M * Bertl sith: are you a dark master? 1109794288 M * eyck probably just wannabe 1109794299 M * eyck runing around and making scary faces 1109794306 M * Bertl guess that's called apprentice then ... 1109794327 M * eyck :) 1109794337 M * sith hehe. 1109794341 M * sith dark lurker maybe, =) 1109794342 M * Bertl eyck: do you read the linux-vserver ml? 1109794371 M * Bertl sith: what is the issue? 1109794397 M * sith Bertl: trying to get RHAS3 update 4 working in a vserver. 1109794406 M * eyck I did, but there was nothing there lately? 1109794416 M * sith but it won't because of the glibc-non-exec-stack patch. 1109794437 M * eyck oh, right, RHA-sth mails... 1109794440 M * sith i'm fine using a non-redhat kernel.. but I can't even get the update4 image to run. 1109794490 A * sith greps through posted irc logs. 1109794571 M * Doener Bertl: regarding http://doener.homeip.net/doener/vserver/diff-2.6.11-rc5-vs1.9.4.9-ng9.1-various.diff i wonder if there was a special reason for not initialising nfvnet in alloc_skb 1109794596 M * Doener (and if the first hunk really fixes a bug or if that was really meant to be skb and not newskb 1109794616 M * Bertl tx Doener! 1109794631 M * Bertl will look at it in a few minutes 1109794639 M * Doener just some small changes, not sure if anything of that is valid :) 1109794703 M * Bertl well, I guess it did improve things for you, right? 1109794746 M * Bertl sith: okay, so the issue is that the 'modified' RH* glibc won't work with a vanilla kernel, right? 1109794779 M * sith Bertl: well, the current (update4) glibc won't work, I didn't modify it, but just did a up2date. 1109794788 M * Bertl eyck, Doener: can anybody tell me what Roderick is talking about? 1109794805 M * sith all redhat versions glibc-2.3.2-95-22 and up won't work AFAICT 1109794812 M * Doener for the first hunk i didn't notice anything, i just noticed it while looking at the code. the latter two seem to make it more logical somehow and it's required for the local communication stuff i play with atm 1109794812 M * Bertl sith: where and how does it fail? 1109794827 M * Doener Bertl: Roderick -> ml? 1109794839 M * Bertl yep, topic cfengine ... 1109794860 M * sith Bertl: its a kernel incompatiblity, it gets something like "Can't map libsomething into non-exec stack". 1109794873 M * eyck roderick? from ML ? 1109794883 M * Bertl yes eyck! 1109794899 M * sith http://list.linux-vserver.org/archive/vserver/msg05173.html - like this. 1109794905 M * Bertl sith: hmm, do you know a kernel patch which is responsible for that? 1109794935 M * eyck cfengine? 1109794960 M * sith Bertl: well, it's the glibc that breaks compatibility with older kernels 1109794966 M * Doener nope, no idea 1109794981 M * Bertl sith: hmm, older like 2.6.11? 1109794997 M * eyck cfengine is evil 1109795003 M * sith so, AFAIK everyone using vserver patches will have this problem. 1109795006 M * eyck or at least useless 1109795015 M * sith maybe it's fixed in the newer 2.6 patches... 1109795029 M * Bertl eyck, Doener: so I'm not the only one who doesn't know what he's talking about? 1109795044 M * sith i'm still using 2.4 patches to match RHAS3 better... maybe that's been fixed somehow in newer versions? 1109795047 M * eyck Bertl: I know cfengine, tried it few years ago, it's useless 1109795070 M * Bertl sith: could you try a 2.6 kernel with vserver and see if that works? 1109795076 M * eyck too complicated etc... 1109795093 M * eyck although using similiar tool for vserver sounds reasonable... 1109795099 M * sith Bertl: yeah, I guess I'll try that. 1109795101 M * sith Bertl: thanks, =) 1109795119 M * eyck because with vservers your servers start multiplying like rabbits, 1109795162 M * eyck but I'm personally looking at Active Directory for solving all those problems. 1109795178 M * Bertl sith: please let us know if that work, at least it's would be some kind of 'workaround' 1109795197 M * eyck anyone fluent with ARM7 assembler? 1109795200 M * sith Bertl: yep, i'd imagine that other people would run into this as well. 1109795226 A * Bertl not fluent, but I guess I can look it up ... 1109795265 M * Bertl hmm, maybe I should decide between /me and /m before ;) 1109795293 M * Bertl *before I write the line, that is .. 1109795336 M * Bertl http://archives.linux-vserver.org/200503/0020.html <-- I'm referring to that email 1109795356 M * Bertl IMHO he is implying that we do use cfengine for whatever? 1109795393 M * Bertl or does he mean some software he uses/wrote and is now using with the new style config? 1109795722 J * Pazzo ~thomas@host130-250.pool8172.interbusiness.it 1109795728 M * Pazzo re 1109795827 M * daniel_hozac Bertl: FYI: Enrico imported a modified version of my dlimit patch into CVS. 1109796017 M * Bertl evening Pazzo! 1109796020 M * Bertl daniel_hozac: excellent! 1109796331 M * Pazzo bertl: another chance to test 1.9.5-rc1: "someone" seems to have a problem with a data corruption bug (fusion mpt driver / megaraid somewhere around 2.6.8 or 2.6.9)... they NEED a bleeding-edge kernel ;-) 1109796376 M * Pazzo what do you consider the best memory split setting for a dual xeon / 4gb ram host? 1109796643 J * CosmicRay ~jgoerzen@2002:4463:7269:1:20e:a6ff:fe66:c5a3 1109796667 M * CosmicRay hiya. I just built a 2.6.11-vs1.9.4 and things are not going well. 1109796673 M * CosmicRay erwin:~# vserver serv1 enter 1109796673 M * CosmicRay Can't set the ipv4 root (Function not implemented) 1109796724 M * CosmicRay now, I haven't done something stupid like forgetting to patch it. 1109796726 M * Doener Bertl: google says: http://www.paul.sladen.org/vserver/archives/200404/0068.html 1109796731 M * Doener * it introduces a completely new configuration format which was designed 1109796731 M * Doener to: 1109796731 M * Doener - be easy changeable with cfengine 1109796763 M * CosmicRay so I'm wondering if anybody else has seen something like this. 1109796793 M * Doener CosmicRay: did you disable linux vserver legacy support? 1109796814 M * CosmicRay Doener: no, I kept it enabled because I'm using old userland 1109796837 M * CosmicRay Doener: and it worked fine that way in 2.6.9-vs1.9.3 1109796845 M * Doener hmm... check uname -a... 1109796847 M * CosmicRay util-vserver 0.30-13 1109796853 M * CosmicRay Linux erwin 2.6.11-vs1.9.4 #1 Wed Mar 2 12:46:40 CST 2005 alpha GNU/Linux 1109796865 M * CosmicRay I verified it's in there and configured 1109796893 M * Doener phone... 1109796899 M * daniel_hozac alpha? cool. 1109796910 M * CosmicRay yep 1109797021 M * Bertl Doener: ah, okay, that explains it ... 1109797064 M * Bertl Pazzo: with 4GB you can take any split, it will not help you to avoid highmem 1109797096 M * Bertl (probably I'd go for 2/2 to get the half of it as lowmem for buffers and such) 1109797126 M * micah has anyone tried 1.9.5-rc1 against the finally released 2.6.11? 1109797127 M * Bertl CosmicRay: can you do some tests for me, maybe the alpha interface is broken somehow? 1109797132 M * CosmicRay berni: sure 1109797144 M * CosmicRay err make that bertl 1109797148 M * micah I'm about to try it now :) 1109797157 M * daniel_hozac Bertl: why is lowmem better than highmem? 1109797170 J * shuri sjnesjd@dsl.speedline209.226.electronicbox.net 1109797183 M * Bertl CosmicRay: http://vserver.13thfloor.at/Stuff/SCRIPT/testme.sh 1109797192 M * shuri hello everybody 1109797197 M * Pazzo thnx bertl! 1109797216 M * daniel_hozac micah: with http://vserver.13thfloor.at/Experimental/patch-2.6.11-vs1.9.5-rc1.diff ? 1109797220 M * Bertl daniel_hozac: well, think of lowmem as of your desk, and highmem are the drawers ... 1109797225 M * Pazzo micah: me too :) 1109797233 M * CosmicRay erwin:/home/jgoerzen/programs/kernel/t# sh testme.sh 1109797233 M * CosmicRay Linux-VServer Test [V0.09] (C) 2003-2005 H.Poetzl 1109797233 M * CosmicRay Can't set the new security context 1109797233 M * CosmicRay : Function not implemented 1109797233 M * CosmicRay chcontext failed! 1109797234 M * CosmicRay Can't set the ipv4 root (Function not implemented) 1109797236 M * CosmicRay chbind failed! 1109797238 M * CosmicRay Linux 2.6.11-vs1.9.4 alpha/0.30/0.30 [J] 1109797240 M * CosmicRay --- 1109797256 M * Bertl you are using 0.30 tools? 1109797259 M * CosmicRay yes 1109797264 M * Bertl please update to 0.30.204 1109797278 M * Bertl the old tools do not know about the non x86 syscalls 1109797284 M * CosmicRay er. easier said than done, I'm using the util-vserver tools in debian 1109797289 M * TheSeer 204? 1109797296 M * CosmicRay well that works fine with 2.6.9-vs1.9.3 1109797296 M * TheSeer i just updated to 203 ;> 1109797298 M * Bertl CosmicRay: throw them away ... 1109797312 M * CosmicRay what would have made them break with 2.6.11-vs1.9.4? 1109797323 M * Bertl CosmicRay: yes, because alpha got an official syscall addres 1109797327 M * CosmicRay ah. 1109797330 M * Bertl before it was the same as with x86 1109797338 M * Pazzo Bertl: and "Allocate 3rd-level pagetables from highmem" -> does this improve performance and / or is it risky? 1109797362 M * Bertl should be fine (well, guess it's still experimental, yes?) 1109797383 M * Pazzo bertl: no 1109797391 M * Bertl then go for it 1109797396 M * Pazzo thnx! 1109797467 M * CosmicRay Bertl: what is the proper url for 0.30.204? 1109797483 M * CosmicRay I'm looking at http://www.13thfloor.at/vserver/project/ and not really finding it 1109797508 Q * mhepp Quit: mhepp caught signal: Autobus error 1109797606 M * CosmicRay closest I can see is 0.30.203 1109797644 M * Doener http://linux-vserver.org/alpha+util-vserver -- link is in the heading 1109797761 M * CosmicRay crap, the debian patch does not apply cleanly to this. 1109797854 M * Bertl CosmicRay: ahem? if you absolutely want a debian version (no price to high) then look for one of the versions the non-maintainers put up on a regular basis ;) 1109797870 Q * ntrs_ Read error: Connection reset by peer 1109797874 J * ntrs_ ~ntrs@Dardeene-68.188.50.87.charter-stl.com 1109797876 M * CosmicRay Bertl: well it is the path of least resistance given the existing setup anyway. 1109797928 M * CosmicRay hmm, you wouldn't happen to have some urls, would ya? 1109798018 M * CosmicRay google is not being the most helpful 1109798020 M * Bertl scan the ml archives, IIRC Snow-Mann (you can easily deduce his real name ;) has posted a few times (where you can find his stuff) 1109798022 M * CosmicRay found just a lot of woody backports 1109798042 M * CosmicRay Snow-Man: hey! ping! 1109798050 M * Bertl well, nowadays everybody and hist dog do woody backports ... 1109798058 M * Bertl *his 1109798065 M * Pazzo *wuff* 1109798127 M * CosmicRay http://kenobi.snowman.net/~sfrost/vserver/ looks promising 1109798178 M * Snow-Man CosmicRay: hrmmmm? 1109798205 M * CosmicRay Snow-Man: two things... 1) I had no idea you were involved with vserver, and 2) I was looking for your util-vserver packages, but I've found them already :-) 1109798208 M * Snow-Man CosmicRay: Yes, that's the vserver 1.9.4 patch against 2.6.10-5, cleaned up to patch happily. 1109798233 M * Snow-Man CosmicRay: I'm more a user of vserver than really involved in any other sense. 1109798241 M * CosmicRay same here 1109798249 M * Snow-Man Though I tend to bitch out the DD who currently 'maintains' the package on occation. 1109798256 M * CosmicRay Snow-Man: me too 1109798257 M * CosmicRay he never listens 1109798270 M * CosmicRay I don't know wtf he persists with packaging that outdated version 1109798272 M * Bertl keep bitching folks! 1109798274 M * Snow-Man CosmicRay: They're not really my util-vserver packages either, someone else did them, I just critiqued them and decided to host them. 1109798291 M * Snow-Man CosmicRay: Because Bertl here refuses to make a 'stable' release. :) 1109798301 M * CosmicRay it ticks me off that vserver is broken out of the box with 2.6 kernels in debian 1109798304 M * Bertl yeah, bad Bertl! 1109798308 M * CosmicRay Bertl: yeah, what's up with that? :-) 1109798336 M * CosmicRay well I jsut submitted a bug report asking for 0.30.204 to be packaged. 1109798358 M * CosmicRay aw crap. 1109798372 M * Snow-Man That doesn't sound good. 1109798385 M * Snow-Man I didn't do it. 1109798386 M * Snow-Man :) 1109798395 M * CosmicRay dpatch apply-all 1109798396 M * CosmicRay applying patch ndim_rollup to ./ ... failed. 1109798396 M * CosmicRay make: *** [patch-stamp] Error 1 1109798409 M * CosmicRay I hate debian packages that use custom patching systems 1109798410 M * Snow-Man wtf is ndim_rollup? 1109798411 M * CosmicRay no offense :-) 1109798415 M * CosmicRay it's in debian/patches 1109798427 M * micah Bertl: I just applied patch-2.6.11-vs1.9.5-rc1.diff to 2.6.11 (the release) everything looks fine except for a few offsets 1109798429 M * Snow-Man Which debian package? 1109798429 M * CosmicRay and it covers a whole bunch of crap 1109798450 M * micah Bertl: ooh, actually, no, there is a FAILED hunk in Makefile.rej 1109798453 M * CosmicRay Snow-Man: you latest 0.3.19x, uupdated to 0.30.204 1109798475 M * micah Bertl: only because the EXTRAVERSION is set to -rc5 in the patch, when it is no longer in 2.6.11 1109798479 M * Snow-Man CosmicRay: Oh, yeah, not my packages, but ok. :) 1109798501 M * Snow-Man The packages I first did when there weren't any others (I don't think) are in old/ 1109798506 M * Snow-Man And are 0.30.195 1109798521 M * Snow-Man I did some critique on the 0.30.196 ones and put them up 'cause I wanted them available somewhere for people. 1109798525 Q * ntrs_ Ping timeout: 480 seconds 1109798529 M * Snow-Man I'm trying to remember who the guy was that made the packages... 1109798535 M * Bertl micah: hmm? 1109798539 M * Snow-Man He was gonna do 204 or whatever when he got a chance. 1109798545 M * CosmicRay -- Kilian Krause Fri, 7 Jan 2005 18:16:31 +0100 1109798550 M * CosmicRay could that be it? 1109798558 M * Snow-Man Yeah, he and another guy working together. 1109798561 M * micah Bertl: I applied that patch to the (today released) 2.6.11 1109798564 M * Bertl http://vserver.13thfloor.at/Experimental/patch-2.6.11-vs1.9.5-rc1.diff 1109798574 M * Bertl (should not contain any -rc5) 1109798579 M * micah oop :h 1109798603 M * micah thats why 1109798622 M * CosmicRay ok, let me just clarify here for a sec. 1109798633 M * CosmicRay 0.30.204 is designed to work with 1.9.x, right/ 1109798642 M * CosmicRay will I need to disable the legacy support in order for this to work for me? 1109798685 M * daniel_hozac the legacy support is _required_ ;) 1109798689 M * CosmicRay ah 1109798691 M * CosmicRay ok then :-) 1109798749 A * CosmicRay finds it amusing that vserver's' configure looks for a fortran compiler 1109798790 M * Bertl btw, we changed the help text for the legacy stuff ;) 1109798809 M * Snow-Man CosmicRay: 0.30.196 will work just fine w/ 1.9.x too, at least that's been my experience with 1.9.4 1109798812 M * Bertl (it should now point out that it .. hmm is somewhat required) 1109798838 M * CosmicRay Snow-Man: apparently there is an alpha-specific issue that I must have 0.30.204 to fix. 1109798855 M * Bertl hmm, not necessarily 1109798859 M * CosmicRay oh? 1109798873 A * CosmicRay may have misunderstood then 1109798879 M * Bertl the tools do support the various platform specific syscalls (when the kernel source is present) 1109798898 M * Bertl and with the alpha util-vserver you also may specify it explicitely 1109798907 M * Bertl just the 'stable' branch does not support it (never has) 1109798943 M * Bertl so except for the many new features and fixes in 0.30.204, 0.30.196 will do 1109798950 M * CosmicRay hmm, so would 0.30.196 work out of the box then, or would I need to give it special switches? 1109798965 A * Snow-Man doesn't know about any 'special switches' he's had to do. 1109798966 Q * ndim Ping timeout: 480 seconds 1109798972 M * Bertl CosmicRay: check it, and let me know what the build time config says ... 1109798976 M * Snow-Man But then, I'm using the .deb, directly... 1109798985 M * Snow-Man I hadn't bothered recompiling it. 1109798992 M * Bertl (no idea if the debian packages work on alpha) 1109799024 M * Snow-Man I wish you wouldn't call things 'alpha' or 'experimental' or whatever. 1109799035 M * Snow-Man By alpha you mean 1.9.x? Or something else this time? 1109799043 M * CosmicRay checking for supported APIs... compat,v11,v13,fscompat,net,oldproc,olduts 1109799043 M * CosmicRay checking for number of syscall 'vserver'... 428/kernel 1109799043 M * CosmicRay checking for syscall(2) invocation method... auto 1109799043 M * CosmicRay checking which syscall(2) invocation works... fast 1109799047 M * CosmicRay is that what you're after, bertl? 1109799049 M * CosmicRay that was with 0.30.196 1109799120 A * Snow-Man can't keep the variety of vserver stuff straight. 1109799127 M * Bertl yep, 428 sounds good 1109799134 M * CosmicRay swet. 1109799136 M * CosmicRay err sweet. 1109799140 M * CosmicRay Snow-Man: me neither 1109799151 M * Snow-Man CosmicRay: Told ya, just use the 0.30.196 debs. :) 1109799159 M * CosmicRay I just harass ola about it since it's clear that his debs suck :-) 1109799162 M * CosmicRay Snow-Man: heh 1109799170 M * CosmicRay of course, I hate it when people do the same to me about cscvs 1109799177 M * Bertl then again, variety is the spice of life ;) 1109799180 J * ndim hun@helena.bawue.de 1109799187 M * CosmicRay but at least I have an excuse there. upstream's repository disappeared :-) 1109799187 M * Snow-Man His debs do suck, it's really annoying. :/ 1109799191 M * CosmicRay heh 1109799192 M * Bertl welcome ndim! 1109799195 M * Snow-Man CosmicRay: Whoops. 1109799207 M * CosmicRay Snow-Man: didn't he maintain postgres in years past? I recall those debs sucking too :-) 1109799220 M * Snow-Man I dunno, but I'm glad he doesn't now. :) 1109799225 M * CosmicRay heh 1109799242 M * CosmicRay ah no, that was oliver elphick. 1109799252 M * Snow-Man Oliver still does 1109799252 M * Bertl just for the record, regarding syscall numbers look here: http://linux-vserver.org/Syscall+Switch+Info 1109799256 M * Snow-Man Well, kind of, it's mostly Martin 1109799261 Q * johnny Ping timeout: 480 seconds 1109799267 M * CosmicRay martin pitt these days 1109799270 M * Snow-Man And I'm on the 'team' except that I havn't done shit but bitch, really. :) 1109799278 M * CosmicRay Snow-Man: preach it, brother :-) 1109799300 M * CosmicRay actually i usually do try to fix things that suck 1109799311 M * CosmicRay but in some cases, it's just beyond my expertise to do so 1109799312 M * Snow-Man CosmicRay: ya damn straight. Pissed me off that postgres had all those extra modules/.debs being built out of the same source when they were different upstreams. 1109799315 J * ntrs ntrs@Dardeene-68.188.50.87.charter-stl.com 1109799323 M * Bertl welcome ntrs! 1109799324 M * CosmicRay Snow-Man: really? that's way broken 1109799339 M * CosmicRay Snow-Man: it just pissed me off that my database broke on every major upgrade 1109799355 M * CosmicRay and I'd have to do the manually-find-old-pg_dump routine 1109799358 M * Snow-Man CosmicRay: Well, I proved to them that it was possible to provide a -dev package that would allow most of those extra postgres modules to build just w/ the -dev package and not in the same source tree. 1109799387 M * Snow-Man CosmicRay: eh, major database upgrades still aren't handled all that well I don't think. 1109799403 M * CosmicRay Snow-Man: been much better for me lately. 1109799408 M * Snow-Man iirc Oliver/Martin are working on a mechanism whereby you can have multiple major versions installed at once. 1109799418 M * CosmicRay yeah, that's sad. 1109799422 A * Snow-Man shrugs. 1109799423 M * CosmicRay I see no point to that at all 1109799429 M * Snow-Man I dunno, that doesn't really bother me. 1109799430 M * CosmicRay libpq can connect to older versions still 1109799456 Q * ntrs Read error: Connection reset by peer 1109799479 M * CosmicRay anyway, I decided that the standard haskell library sucked in some big ways, so I wrote my own (MissingH) to provide the missing features 1109799493 M * CosmicRay and I wrote MissingPy to provide a binding to Python to provide the missing features I didn't want to re-implement myself :-) 1109799584 M * CosmicRay bah 1109799589 M * Snow-Man haha 1109799591 M * Snow-Man Fun. 1109799614 M * Snow-Man Gotta go. :) 1109799620 M * CosmicRay now it thinks all my vservers are in /var/lib/vservers instead of /home. 1109799625 M * CosmicRay I probably did some evil hack to fix that last time 1109799628 M * CosmicRay err /home/vservers. 1109799677 M * CosmicRay is it safe to symlink /var/lib/vserevrs to /home/vservers? 1109799810 A * CosmicRay just rebuilds with a new arg to configure 1109799881 M * Pazzo CosmicRay: with alpha-utils new config style there is a /etc/vservers/.defaults/vdirbase symlink... 1109799916 M * CosmicRay I noticed that, but it didn't seem to make any different 1109799919 M * CosmicRay err difference 1109799976 M * Doener hm, replacing that symlink _should_ make a difference... 1109800043 M * Doener and now that you mention that... Hollow: i guess that symlink is not cfg-protected in the .204 ebuild, at least my symlink was pointing to the default location after upgrading 1109800043 M * Pazzo CosmicRay: strange... sorry, I never tried it out... I was always using my own debs (similar to Kilian Krause / snow-man / ... - ones) and built them (not debian-like) to use /vservers 1109800060 M * CosmicRay Pazzo: no problem 1109800078 M * CosmicRay erwin:/etc/vservers/.defaults# ls -l 1109800079 M * CosmicRay total 0 1109800079 M * CosmicRay drwxr-xr-x 2 root root 48 Mar 2 15:36 apps 1109800079 M * CosmicRay drwxr-xr-x 2 root root 48 Mar 2 15:36 files 1109800079 M * CosmicRay lrwxrwxrwx 1 root root 21 Mar 2 15:37 run.rev -> /var/run/vservers.rev 1109800079 M * CosmicRay lrwxrwxrwx 1 root root 14 Mar 2 15:39 vdirbase -> /home/vservers 1109800081 M * CosmicRay erwin:/etc/vservers/.defaults# /etc/init.d/vservers start 1109800085 M * CosmicRay Starting the virtual servers 1109800087 M * CosmicRay WARNING: can not find configuration, assuming legacy method 1109800089 M * CosmicRay No directory for this vserver: /var/lib/vservers/serv1 1109800176 J * Nik ~Nik@cable-153-130.online.bg 1109800183 M * Nik hi all 1109800233 M * CosmicRay hey hey! it works. 1109800246 M * CosmicRay thanks everyone for your help 1109800249 M * CosmicRay and sorry about bitching so much 1109800259 M * CosmicRay vserver is great, really 1109800273 M * CosmicRay and y'all are doing great things with it 1109800300 M * Doener what did you change to make it work? 1109800370 M * CosmicRay updated util-vserver to 0.30.196, and adjusted the parameter to configure to point to my correct local vservers path 1109800403 M * Doener ah, now i see... legacy config. yep, the tools for those use a hardcoded path 1109800846 M * Bertl hey Nik! 1109800911 Q * Nik Ping timeout: 480 seconds 1109801076 J * hillct ~hillct@65.76.245.91 1109801079 M * Doener guess you scared him ;) 1109801084 M * CosmicRay heh 1109801101 M * Bertl probably, let's try again! 1109801113 M * Bertl hello hillct! 1109801117 M * hillct hey 1109801206 M * hillct I'm still messing with immutable link invert 1109801240 M * hillct I rewrote most of vskel.pl to utilize the modern configuration mechanism 1109801256 M * Bertl you like that one, right? 1109801286 M * Doener hillct: what does vskel do? a patch for that one was my first post to the ml, but i don't remember what it does anymore ;) 1109801288 M * hillct I'm not familiar wit enough different mechanisms to have an informed opinion 1109801323 M * hillct vskel.pl basically takes a server filesystem and beaks it up into global, local and static items 1109801334 M * hillct the global items are hard linked 1109801340 M * hillct the loal items are copied 1109801356 M * Bertl and the static ones? thrown away? 1109801357 M * hillct and the static items are key-replaced 1109801371 M * hillct resolv.conf etc 1109801387 M * Bertl ah, so basically generic template rewriting 1109801387 M * hillct key replaced for IP addresses hostnames etc 1109801391 M * hillct yes 1109801398 M * hillct although 1109801419 M * hillct I'm having a problem with immutable link invert still. 1109801421 M * hillct sort of 1109801438 M * hillct I use 1109801467 M * hillct setattr --imunlink for the global files in the skeleton directory 1109801489 M * hillct but then when I go to create a vserver from this skeleton, it fails to create the links 1109801530 M * Bertl --iunlink I hope ... 1109801534 M * hillct yes 1109801536 M * hillct typo 1109801590 M * Bertl and you mean you fail to create new links to such a file on the host? 1109801601 M * hillct yes 1109801615 M * hillct is this a FS attribute support problem 1109801623 M * hillct there was discussion of a patch for reiserfs 1109801624 M * Bertl okay, let me try that .. sec 1109801630 A * hillct uses XFS 1109801644 M * hillct is it not native? 1109801661 M * Bertl what natvie? 1109801683 M * hillct support for the immutable link invert bit 1109801706 M * hillct there was a reiserfs patch. is a patch required for XFS? 1109801734 M * Bertl well, the main thing is that filesystems like xfs usually implement such things like xattrs as they 'think' they should work (telling from hear-say) 1109801739 M * hillct I looked through kernel options looking for a way to enable/disable the attribute 1109801759 M * Bertl for example reiserfs need a mount option IIRC to enable xattribs 1109801769 M * hillct ah 1109801792 M * Bertl and each fs (except for ext2/ext3) uses their own 'rules' how to handle them ... rather nasty in the kernel ... 1109801828 M * Bertl but it might also be that we removed an existing feature (maybe accidentially or on purpose) and nobody complained yet ;) 1109801858 A * hillct logs in to get version info 1109801867 M * hillct using utils 1.9.4 I think 1109801948 M * Bertl s/utils/kernle patch/ 1109801982 M * hillct patch-2.6.10-vs1.9.3.17 1109801990 Q * erwan_ho Remote host closed the connection 1109802047 A * Bertl is compiling xfs support into his testkernel 1109802086 M * hillct I may be able to get access to a 1.9.3.14 installation to compare 1109802098 M * hillct to see if it broke recently 1109802109 M * hillct darn 1109802120 M * hillct can't get to it from here 1109802147 M * Bertl np, if it is broken, we'll fix it ;) 1109802183 M * hillct is there something I should do to confirm that It's not just my installation. I'm really new to this 1109802196 M * hillct ? 1109802215 M * Bertl okay, first the obvious check: does it work on ext2/ext3 (doing here now, if you could do that with a loopback or spare partition, it would be great) 1109802262 M * hillct hmm 1109802268 A * hillct looks for space 1109802391 M * Bertl # touch /vservers/test 1109802391 M * Bertl # setattr --iunlink /vservers/test 1109802391 M * Bertl # showattr /vservers/test 1109802391 M * Bertl ----UI- /vservers/test 1109802391 M * Bertl # ln /vservers/test /vservers/test2 1109802394 M * Bertl # showattr /vservers/test* 1109802396 M * Bertl ----UI- /vservers/test 1109802399 M * Bertl ----UI- /vservers/test2 1109802406 M * Bertl (this is with ext2 and it seems to work just fine) 1109802448 A * hillct locates loopback fs docs 1109802455 M * hillct this will take a moment 1109802503 M * Bertl # touch /vservers/test 1109802503 M * Bertl # setattr --iunlink /vservers/test 1109802503 M * Bertl # showattr /vservers/test* 1109802503 M * Bertl ----uI- /vservers/test 1109802514 M * Bertl (this is with xfs, which doesn't look so good) 1109802652 M * Bertl interestingly a remount reveals the 'U' attribute as set 1109802673 M * Bertl Doener: didn't we do a patch to fix something like that? maybe that was lost or just for a different fs? 1109802705 M * Doener not that i remember 1109802737 M * Bertl hmm, let me scan my mailbox, I'm pretty sure I remember seeing something like that before ... 1109802820 M * hillct on my system under ext2: 1109802822 M * hillct [root@laptop temp-vserver]# touch testfile 1109802822 M * hillct [root@laptop temp-vserver]# setattr --iunlink testfile 1109802822 M * hillct [root@laptop temp-vserver]# showattr testfile 1109802822 M * hillct ----UI- testfile 1109802822 M * hillct [root@laptop temp-vserver]# 1109803020 M * hillct it does seem to operte properly under ext2 1109803022 M * hillct here 1109803087 M * Bertl yeah, I can reproduce the issue here, and I have a good idea what goes wrong .. when we fix it, can you easily test it? 1109803188 M * hillct sure 1109803210 M * hillct I planned to repartition and do my final install in the next few days 1109803227 M * hillct I can redo the kernel any time 1109803285 M * Bertl Doener: forwarded you an email ;) 1109803346 M * Doener ah, now i remember 1109803396 M * hillct I don't know enough about the attribute notation I guess. What's the lovercase vs upper case? 1109803397 M * Pazzo @all: good night! 1109803404 M * Bertl I'd say xfs and maybe reiser is missing something like that ... 1109803409 M * hillct g'night Pazzo 1109803410 M * Bertl night Pazzo! 1109803432 M * Bertl lowercase means 'available' but unset, upper case means available and set 1109803442 M * hillct ah 1109803453 M * Pazzo Bertl: 2 different Debian packages with 2.6.11-vs1.9.5-rc1 are ready, I'll start testing tomorrow... 1109803456 M * Pazzo cya 1109803459 Q * Pazzo Quit: ... 1109803464 M * hillct oh 1109803485 M * hillct there's a slight change needed in the spec file for 1.9.x utils 1109803505 M * hillct it uses chattr to setup the /vservers barrier 1109803513 M * hillct should use setarr 1109803521 M * hillct in the post script 1109803530 M * Bertl hmm, I doubt that, btw, tools are 0.30.204 ... 1109803556 M * hillct hmm 1109803580 M * hillct tools vs utils? what do tools provide as distinct from vserver-utils? 1109803603 A * hillct grabs notes 1109803608 M * Bertl well, you are using debian, right? 1109803617 M * hillct fedora 2 1109803633 M * Bertl hmm, interesting, fc2 should not know this confusion ... 1109803652 M * Bertl i.e. there is util-vserver ('THE TOOLS') and the kernel patch ... 1109803661 M * hillct K 1109803729 M * hillct damnit. connection just humg up on me 1109803730 M * hillct grr 1109803758 M * hillct 1109803761 M * hillct [root@laptop x86_64]# rpm -ivh util-vserver-0.30.209-0.x86_64.rpm 1109803761 M * hillct util-vserver-core-0.30.209-0.x86_64.rpm util-vserver-lib-0.30.209-0.x86_64.rpm 1109803761 M * hillct util-vserver-sysv-0.30.209-0.x86_64.rpm 1109803761 M * hillct 1109803761 M * hillct Preparing... ########################################### [100%] 1109803763 M * hillct 1:util-vserver-core ########################################### [ 25%] 1109803763 M * hillct 2:util-vserver-lib ########################################### [ 50%] 1109803765 M * hillct 3:util-vserver-sysv ########################################### [ 75%] 1109803765 M * hillct 4:util-vserver ########################################### [100%] 1109803767 M * hillct 1109803784 M * Bertl yes, that's util-vserver (separated in packages) 1109803792 M * hillct yah 1109803816 M * hillct the command that generates the error is chattr -it /vservers 1109803828 M * hillct shouldn't it be: 1109803832 M * Bertl 209 looks a little weird, but maybe we missed some releases (o,o) 1109803837 M * hillct setattr --barrier /vservers 1109803923 M * hillct I'm trying to understand the --unify functionality in the main vserver script 1109803937 M * Bertl hmm, hmm, where do you have those rpms from? 1109803950 M * hillct is the idea to copy a reference vserver to create a new one then unify afterward, to get the hard links? 1109803954 M * Bertl (because the developer is still at 0.30.204 ;) 1109803961 M * hillct uh 1109803964 M * hillct rpms... 1109803966 M * hillct hmm 1109803968 M * hillct hang on 1109803973 M * Bertl so it's like rpms from the future ;) 1109804009 M * Bertl and of course, if we _know_ we are going to make those mistakes, there's no point in trying to avoid them, right? ;) 1109804015 M * hillct built them from a tarball off 13thfloor 1109804030 M * Bertl no way, I have no 0.30.209 there! 1109804039 A * hillct verifies 1109804069 M * hillct ls 1109804073 M * hillct d'oh 1109804083 Q * tchan Quit: leaving 1109804129 M * hillct I have the src rpm still 1109804180 M * Bertl well, do you know where you got it from (i.e. url)? 1109804212 M * hillct putting it up for you to retrieve 1109804219 M * hillct then I'll check the spec file 1109804248 M * hillct http://www.pinnacledigital.com/SRPMS/util-vserver-0.29.209-0.src.rpm 1109804259 M * hillct this is the src rpm I built from the tarball 1109804313 M * Bertl ah, that's pretty old, it's 0.29.209 1109804323 M * Bertl we are at 0.30.204 ;) 1109804326 M * hillct http://savannah.nongnu.org/projects/util-vserver/ 1109804328 M * hillct ah 1109804340 M * hillct heh 1109804344 M * hillct one version off 1109804348 M * hillct that's pretty bad 1109804351 M * Bertl almost a year later ;) 1109804367 M * hillct how the hell did that happen 1109804372 A * hillct knows better 1109804376 M * Bertl LOL 1109804382 M * hillct I swear!!! 1109804384 M * hillct I do! 1109804407 M * Bertl yeah, sure, irc log proves that ;) 1109804417 M * hillct somehow I did the increment in my head 1109804451 M * Bertl I still find it interesting that you have listed fictional rpms ;) 1109804453 M * hillct but I did have the right kernel patches 1109804491 A * hillct does another sanity check 1109804557 A * hillct kicks self 1109804570 A * hillct builds new util-vserver 1109804588 M * hillct guess that may explain this behavior... 1109804599 M * Bertl yeah, pretty much ... 1109805027 M * hillct so the new spec file builds everything as a single RPM? 1109805048 M * Bertl no, it is stil sepaated in logical units 1109805057 M * jd86 how do i tell the gateway of a running linux box where theres no config? 1109805086 M * hillct just not the same logical units 1109805092 M * hillct I only got util-vserver-0.30-0.x86_64.rpm 1109805092 M * hillct util-vserver-debuginfo-0.30-0.x86_64.rpm 1109805092 M * hillct util-vserver-linuxconf-0.30-0.x86_64.rpm 1109805105 A * hillct checks the spec 1109805140 M * hillct batteries about to bie 1109805143 M * hillct die 1109805144 M * Bertl those are the stable tools (0.30 vs 0.30.204 -- note the difference;) 1109805173 Q * prae Quit: Pwet 1109805226 Q * CosmicRay Quit: Client exiting 1109805256 M * hillct I've become a cliche 1109805264 M * hillct sitting in starbucks working away 1109805286 M * hillct K 1109805301 M * hillct you recommend the development tools release? 1109805313 M * Bertl for the development branch, yes ;) 1109805321 M * hillct yah 1109805429 M * hillct should I be pulling it out of CVS or is there a dev tools release tarball? 1109805439 M * Bertl http://www-user.tu-chemnitz.de/~ensc/util-vserver/alpha/ 1109805443 M * hillct ah 1109805459 M * hillct is that linked off linux-vercvers.com? 1109805461 M * hillct missed it 1109805471 M * hillct not on savanah either 1109805491 M * Bertl yes, and the base url is linux-vserver.org ;) 1109805537 M * hillct K 1109805543 A * hillct will look again later 1109805553 M * hillct perhaps I just suck 1109805569 J * ccooke ~ccooke@spc1-walt1-4-0-cust221.asfd.broadband.ntl.com 1109805580 M * Bertl hillct: np, we will shape you up ;) 1109805585 M * Bertl welcome ccooke! 1109805592 M * ccooke Evening 1109805606 M * Bertl Doener: I absolutely fail to see where xfs goes wrong :/ 1109805749 M * hillct er, so you were uting the recent toolset and still saw the problem, so it does exist... 1109805759 A * hillct feels he doesn't suck quite so much 1109805765 M * Bertl yes the issue is confirmed! 1109805781 M * hillct forgime me if I feel good about that ;) 1109805820 A * Doener tests his luck 1109806947 M * Bertl Doener: hmm, look at xfs_dic2xflags() and XFS_DIFLAG_ANY 1109807062 Q * nox Ping timeout: 480 seconds 1109807090 J * nox ~nox@213.39.207.160 1109807094 M * Doener yep, that would make sense 1109807102 M * Bertl okay, trying now ... 1109807131 M * Bertl could you check, it looks to me as if that wasn't there a few kernel releases ago ... 1109807145 M * Bertl the xfs_dic2xflags() I mean ... 1109807312 M * Doener was in 2.6.9, got no olders kernels at hand 1109807345 M * Bertl hmm, maybe I'm just confused ... 1109807388 M * Bertl XFS mounting filesystem hdb1 1109807388 M * Bertl # touch /vservers/test 1109807388 M * Bertl # setattr --iunlink /vservers/test 1109807388 M * Bertl # showattr /vservers/test* 1109807388 M * Bertl ----UI- /vservers/test 1109807390 M * Bertl # ln /vservers/test /vservers/test2 1109807393 M * Bertl # showattr /vservers/test* 1109807394 M * Doener was in 2.6.7, too 1109807395 M * Bertl ----UI- /vservers/test 1109807398 M * Bertl ----UI- /vservers/test2 1109807400 M * Bertl so I guess that is fine now *G* 1109807420 M * Bertl just to make sure let's test reiserfs too ;) 1109807461 Q * nox Quit: I shouldn't really be here - dircproxy 1.0.5 1109807478 J * tchan ~tchan@c-24-13-81-164.client.comcast.net 1109807498 J * jsambrook ~jsambrook@aelfric.plus.com 1109807501 M * Bertl okay, also works fine with reiser ... 1109807507 M * Bertl welcome tchan, jsambrook! 1109807537 P * jsambrook 1109807637 J * Nik ~Nik@cable-153-130.online.bg 1109807643 M * Bertl hmm, but reiser forgets on unmount ... 1109807667 M * Bertl (but I guess that's the dubious reiser option .. checking now) 1109807668 J * ntrs ntrs@Dardeene-68.188.50.87.charter-stl.com 1109807928 J * ntrs_ ntrs@Dardeene-68.188.50.87.charter-stl.com 1109807929 Q * ntrs Read error: Connection reset by peer 1109807986 M * Bertl ah, I love those intuitive mount options 1109807994 M * Bertl {"attrs", .setmask = 1<