1111969650 M * Bertl Doener: still awake and interested in testing something new, kernel wise? 1111969665 M * Doener yes, what is it? 1111969670 M * Bertl sec 1111969805 M * Bertl http://vserver.13thfloor.at/Experimental/delta-2.6.11.6-cow03.diff 1111969826 M * Bertl it's a dirty hack, based on Jörns generic_sendpage 1111969846 M * Bertl it basically turns all hard links into cow links 1111969864 M * Bertl (just for the open case, not chmod or chown yet) 1111969902 M * Bertl you ahve to ignore the warnings (or clean them up ;) 1111969937 M * Bertl (includes the qemu patch) 1111970212 M * yarihm n8 everyone 1111970220 Q * yarihm Quit: Leaving 1111971196 M * Doener Bertl: please enlighten me on cow_break_link... i don't really get the vfs_rename there... 1111971244 M * Bertl np, it took some time to figure the various parts out ... 1111971246 M * Doener we create a copy of the file at /tmp/.cow and then replace the original file with the copy? 1111971256 M * Bertl yep, precisely 1111971262 M * Bertl ah, one detail for this hack 1111971279 M * Bertl you ahve to put your liked file into /tmp/ 1111971302 M * Bertl (i.e. the to file must be in the same dir) 1111971312 M * Doener so what do we gain by replacing a file with a copy of itself? 1111971324 M * Bertl it breaks the links 1111971333 M * Bertl example: 1111971338 M * Bertl echo "test" >/tmp/x 1111971339 M * Bertl ln /tmp/x /tmp/y 1111971343 M * Bertl echo "hansi" >>/tmp/y 1111971349 M * Bertl cat /tmp/y 1111971350 M * Bertl test 1111971350 M * Bertl hansi 1111971363 M * Bertl while /tmp/x still contains test 1111971367 M * Doener ah, now i see... 1111971409 M * Doener i obviously missed the hardlink part ;) 1111971449 M * Bertl it still needs a bunch of error handling if()s and the 'to' has to be generated on the fly 1111971463 M * Bertl it's basically just a proof of concept 1111971643 M * Doener is there any need for the intermediate file at all? i don't know the vfs stuff well, but IIRC in userspace you can open a file, rm it and still work with it, it's really deleted once noone uses it anymore... 1111971654 Q * lilo Quit: brb 1111971668 M * Bertl yeah, right, but there is one problem here 1111971687 M * Bertl the sendfile will take some time to 'copy' the stuff, and it might even fail 1111971701 M * Bertl during this time, we want to keep the original in place 1111974001 M * Doener ok, it seems to work... anything in particular you want me to bash on? 1111974119 M * Bertl well, I need a few ideas regarding how to build (and probably check) the temp name 1111974139 M * Bertl and what special cases to expect ... 1111974169 M * Bertl but nothing specific 1111974287 M * Beirdo Hmmm 1111974316 M * Beirdo I had a vserver build hang mid-install (FC1)... how can I kill the context completely? 1111974379 M * Bertl vkill 1111974471 M * Beirdo that got it. Too obvious, thanks :) 1111974559 J * lilo ~lilo@lilo.usercloak.oftc.net 1111974607 M * Bertl you're welcome 1111974669 M * Beirdo hmph 1111974682 M * Beirdo it sits there and hangs on the dev rpm 1111974683 M * Beirdo :( 1111974813 M * nox Bertl: i gave i vserver CAP_SYS_RAWIO and tty but still get IOPL not permitted 1111974977 M * nox when i try to start X 1111975095 M * Doener nox: strace available? 1111975145 M * Bertl iopl is memory mapping 1111975203 M * Bertl but yes strace would be interesting 1111975299 M * nox sec 1111975571 M * nox http://linuxkun.de/vserver/startx-vps-errors.txt 1111975830 M * Bertl well, I see no indication of anything getting rejected .. 1111975831 M * Bertl try strace -fF X 1111975902 M * nox ok 1111976125 M * nox http://linuxkun.de/vserver/strac-fF_X.txt 1111976169 M * nox i buy an e *g* 1111976408 M * Bertl well, no idea, don't we have a problematic programs entry for X? 1111976449 M * nox Linux-Vserver+FAQ is some info but didn´t help 1111976529 M * nox well try to continue othertime. 1111976573 M * nox ...and include collected knowledge to docs (: 1111976597 M * Bertl k 1111976776 M * Doener hm, that strace just says: hey, no /dev/tty0 (which is a symlink to /dev/vc/0 on my box), but i want one!... guess you only made tty, but not vc/0 ? 1111976859 M * nox oh that right ... i try with 1111977014 M * nox hmm don´t have any vc/ only vcsXX 1111977030 M * nox crw------- 1 root tty 7, 0 2005-02-26 07:39 /dev/vcs0 1111977044 M * nox crw------- 1 root tty 4, 0 2005-03-25 23:50 /dev/tty0 1111977154 M * Doener well, tty0 is not a symlink in your case, so just creating/copying tty0 should be fine 1111977203 M * nox well i had tty0 allready and vcs didn´t help 1111977621 M * Doener why does it say -ENOENT then when trying to access tty0? 1111977883 M * nox open("/dev/tty0", O_WRONLY) = 4 1111977917 M * nox hmm evt was strace without tty (tried both) but result is same anyway 1111978246 M * Doener hm, could you provide a strace of a run where tty0 is available? 1111978406 M * nox http://linuxkun.de/vserver/strace-fF_X+tty.txt 1111978464 M * nox sorry Doener i gonna go ... thx 4 help 1111978470 M * Doener cya! 1111978500 M * nox night chan 1111978777 M * Bertl night! 1111978942 M * Doener hm... only missing CAP_SYS_RAWIO should lead to -EPERM... 1111979372 M * Bertl yep, agreed, we'll check when he get's back 1111985144 M * Bertl okay, folks, I'm off to bed now ... 1111985156 M * Bertl have a good whatever everyone ... cya later ... 1111985160 M * Bertl night Doener! 1111985166 N * Bertl Bertl_zZ 1111985174 M * Doener night Bertl_zZ! 1111989153 J * grecea ~grecea@h-195-22-237-74.mdl.net 1111989372 J * monrad ~monrad@213083190130.sonofon.dk 1111997136 J * sebd ~sebd@lesdeveloppementsdurables.org 1112003035 Q * sannes Read error: Connection reset by peer 1112004301 Q * ruut Ping timeout: 480 seconds 1112004395 Q * lilo Quit: bbiab 1112004807 J * lilo ~lilo@lilo.usercloak.oftc.net 1112006902 J * yarihm ~yarihm@80-218-3-115.dclient.hispeed.ch 1112008239 Q * Hollow Remote host closed the connection 1112008815 J * erwan_ho ~erwan@lns-vlq-39f-81-56-133-136.adsl.proxad.net 1112009417 Q * monrad Quit: Leaving 1112009842 J * sannes ~ace@home.skarby.no 1112010075 J * Hollow ~Hollow@home.xnull.de 1112012440 M * maharaja re 1112012473 M * maharaja compiling a new vserver system - is 2.4 still the way to go? 1112012534 M * eyck interesting question, 1112012546 M * eyck I'm on 2.4, but it seems like Bertl made great progress on new vserver on 2.6.x 1112012640 M * daniel_hozac if you don't want the features of 2.6, i'd say go with 2.4. 1112012640 M * maharaja i recently tried to convert a 2.6.x non vserver system to 2.6.x vserver 1112012645 M * maharaja on a remote box 1112012649 M * maharaja broke everything ;) 1112012658 M * maharaja and i was under pressure so i simply switched back 1112012661 M * maharaja and now got a new machine 1112012668 M * maharaja ok, ill check the 2.6 features 1112012777 M * maharaja hell - there are a lot of things i do not know :) 1112012797 M * maharaja where do i look if i need an explanation of the features? 1112012800 M * maharaja like 1112012808 M * maharaja whats "Hashed RCU for xid/nid/dlim" 1112012821 M * maharaja or "NFS based File Tagging" 1112012828 J * Vudumen_ vudumen@perverz.hu 1112012838 J * tchan_ ~tchan@c-24-13-81-164.client.comcast.net 1112012975 J * brc_ bruce@200165239109.user.veloxzone.com.br 1112012979 J * dmnd wouter@matrix.fatal-exception.nl 1112012982 M * dmnd hello 1112013017 M * dmnd someone who knows if there's a redhat 7.2 or 7.3 guest image available ? 1112013027 M * dmnd i know about the RH9.0 1112013030 M * eyck khehe, http://dump.iof.ru/data/2005_03_28_iclub_nsu_ru_svetotat_pic_cover_neosmopoliten.jpg 1112013030 Q * Vudumen Ping timeout: 480 seconds 1112013030 Q * tchan Ping timeout: 480 seconds 1112013030 Q * brc Ping timeout: 480 seconds 1112013053 Q * DuckKing oxygen.oftc.net orion.oftc.net 1112013053 Q * ndim oxygen.oftc.net orion.oftc.net 1112013053 Q * Steven| oxygen.oftc.net orion.oftc.net 1112013053 Q * jd86 oxygen.oftc.net orion.oftc.net 1112013053 Q * Jason357 oxygen.oftc.net orion.oftc.net 1112013053 Q * eyck oxygen.oftc.net orion.oftc.net 1112013053 Q * click oxygen.oftc.net orion.oftc.net 1112013053 Q * Seraph oxygen.oftc.net orion.oftc.net 1112013053 Q * mikegrb oxygen.oftc.net orion.oftc.net 1112013117 J * DuckKing ~Duck@dyn-83-157-151-194.ppp.tiscali.fr 1112013117 J * ndim hun@helena.bawue.de 1112013117 J * Steven| ~dsf@66-215-207-215.riv-eres.charterpipeline.net 1112013117 J * jd86 ~jim@ip68-9-97-23.ri.ri.cox.net 1112013117 J * Jason357 ~m00@67.159.26.120 1112013117 J * eyck eyck@81.219.64.71 1112013117 J * mikegrb ~michael@mikegrb.netop.oftc.net 1112013117 J * click click@dsl-static-122-208.aal.tiscali.no 1112013117 J * Seraph kk@projects.verfaction.de 1112013119 M * maharaja wb :) 1112013138 M * maharaja eyck: do you know where i can read about what feature does what in the system? 1112013177 Q * Beirdo Ping timeout: 480 seconds 1112013213 J * Beirdo ~gjhurlbu@beirdo.usercloak.oftc.net 1112013626 M * eyck maharaja: I don't understand the question. 1112013809 M * maharaja i dont know what "hased rcu for xid/nid/dlim" means - so i cannot judge if it will benefit me 1112013819 M * maharaja therefore, i'm searching for a place to read about that feature 1112013825 M * maharaja or "nfs based file tagging" 1112013843 M * maharaja i've only been into the 1.3.x releases yet 1112014026 M * maharaja eyck: better now? :) 1112014158 Q * erwan_ho Read error: Operation timed out 1112014163 M * eyck hm, look in wiki and ML archives 1112014174 M * eyck I don't know what hashed rcu does.. 1112014270 M * maharaja mhm 1112014275 M * maharaja i fail to patch 2.4.29 1112014276 M * maharaja damn :) 1112014284 M * maharaja can't find file to patch at input line 3100 1112014284 M * maharaja Perhaps you used the wrong -p or --strip option? 1112014284 M * maharaja The text leading up to this was: 1112014371 M * maharaja ah well - ill give 1.9 a shot 1112014381 M * maharaja as 1.3.9 does not cleanlly apply 1112014421 M * eyck hmm, I think 1.3 is not actively maintained, 1112014438 M * eyck you should go either for 1.2.10+2.4.x, or for 2.6.x 1112014485 M * eyck why would you want 1.3.? it's experimental branch for stable kernel. this makes no sense;) 1112014566 M * maharaja well - i used 1.3 since a long time :) 1112014585 M * maharaja habbits are hard to change... 1112014635 M * maharaja trying my luck with 2.6.11.6-grsec-vs 1112014722 M * eyck btw, who's responsible for grsec + vs ? 1112014809 M * maharaja trying it myself 1112014820 M * maharaja and i get a hell lot of rejects :) 1112014850 M * maharaja a older patch may be found at 1112014850 M * maharaja http://team.lea-linux.org/bgigon/vserver/ 1112014867 M * maharaja but its not the newest grsec so its vulnerable 1112014882 M * maharaja ah, damn 1112015917 J * ruut ~ruut@p5488C552.dip0.t-ipconnect.de 1112018703 M * alexx eyck, prae make a grsec + vs patch 1112018782 M * alexx eyck, but his last version isn't online now :/ 1112018820 M * alexx ho, i found new location : http://studip.uni-greifswald.de/~heim/patches/vserver-grsecurity/ ;) 1112018849 M * DaCa maharaja: http://www.firehead.org/~jeffrey/linux-vserver/ has 2.1.3 for 2.4.29 1112018884 M * DaCa maharaja: which isnt latest either, but at least it doesnt have the PaX vuln. 1112021299 M * Hollow alexx: ~heim's patches are currently not working afaik 1112021321 M * alexx ok, sorry ... i haven't tested myself :/ 1112021335 M * Hollow i played around with him yesterday.. 1112021352 M * Hollow but as soon as he's back we'll take another look at it 1112021528 M * Hollow alexx: are you using grsec in produtivity? 1112021534 M * alexx nop 1112021556 M * alexx actualy, i use 2.6.11.5-vs1.9.5 1112021571 M * Hollow but you would like tot try? 1112021602 M * alexx i've try an old relase on my test machine 1112021637 M * Hollow cause i don't know grsec, and i don't like to learn, so i'd need someone to test if the patches work as expected 1112021655 M * alexx erf :) 1112021665 M * Hollow erf? 1112021667 M * alexx i'm not a grsec specialist myself 1112021670 M * Hollow heh 1112021692 M * alexx i've compiled with it, launch some test vserver, it work, ... but after that ... 1112021720 M * alexx only a production use can be real test :) 1112021764 M * Hollow that's where we are at too ;) 1112024478 M * Hollow alexx: this one should work according to DaPhreak (~heim) http://studip.uni-greifswald.de/~heim/patches/vserver-grsecurity/linux-2.6.11.6-vs1.9.5-grsec-2.1.4.diff.bz2 1112024834 M * alexx k 1112025092 J * micah_ micah@micha.hampshire.edu 1112025467 J * micah__ micah@micha.hampshire.edu 1112025513 Q * micah Ping timeout: 480 seconds 1112025887 Q * micah_ Ping timeout: 480 seconds 1112025967 J * micah micah@micha.hampshire.edu 1112026388 Q * micah__ Ping timeout: 480 seconds 1112026411 N * Bertl_zZ Bertl 1112026425 M * Bertl morning folks! 1112026486 M * eyck morning 1112026519 M * nox re Bertl 1112026582 M * Bertl hey eyck! nox! 1112026656 M * Bertl nox: we came to the conclusion (yesterday) that we are not convinced that you have the required cap inside your vserver (for X) 1112026699 M * nox machine is booting ... any test 2 verify ? 1112026710 M * Doener cat /proc/self/status 1112026716 M * Doener (inside the vserver) 1112026719 M * Doener morning Bertl! 1112026721 M * nox ok sec 1112026738 M * Bertl morning Doener! 1112026987 M * nox omg it was to late yesterday *blush* 1112026993 M * eyck nox is doing X inside vserver? 1112027001 M * nox cat /etc/vservers/X/bcapabilties <---- f*** 1112027044 M * Hollow morning Bertl 1112027147 M * Hollow i rediffed ngnet 9.4 against 2.6.11.6-vs1.9.5 today and it will be optionally included in the ebuild 1112027357 M * Bertl morning Hollow! 1112027405 M * nox Bertl: Doener i´m so sorry ... he wants /dev/mem also but know it works. 1112027461 M * Bertl excellent! ;) 1112027500 M * Doener nox: no problem, shit happens ;) 1112027568 M * nox *blush* 1112029035 J * monrad ~monrad@213083190130.sonofon.dk 1112029170 M * Bertl welcome monrad! 1112029217 M * monrad evening 1112029465 M * Bertl Doener: so what do you think about the cow stuff, can we add that in an useful way without breaking too much? 1112029477 M * Bertl (once it has been cleaned up of course) 1112029520 M * Bertl the IS_COW_LINK() would then become identical to checking the IUNLINK flag ... 1112029555 M * Bertl and in a second strp we could show COW marked links with a link count of 1 1112029556 M * Doener what came to mind was: file X is cow'ed when process A has file X opened (reading) and process B writes to file X, will process A see what B writes? 1112029593 M * Bertl no, but that's not guaranteed with normal files either, IIRC 1112029664 M * Bertl (see the various tricks tail and friends do with files) 1112029724 M * Doener hm... let's try if tail works as it should then ;) 1112029966 M * Doener no, fails 1112030015 M * Doener *lol* this is funny now ;) 1112030033 M * Doener touch test; ln test test2; tail -f test 1112030050 M * Doener in another shell: echo 123 >> test [nothing happens] 1112030062 M * Doener then: echo 123 >> test2 [tail outputs 123] 1112030115 M * Bertl hmm, right ... 1112030208 M * Bertl guess we need the two level inode approach Jörn suggested to handle those cases correctly 1112031127 M * Bertl not sure that would help though ... 1112031155 J * psg ~root@zeb-att-nat.cooldog.com 1112031163 M * Bertl welcome psg! 1112031179 M * psg Hello. Herbert, are you around? 1112031188 M * Bertl yes, I am ;) 1112031194 M * psg Aha! 1112031215 M * psg Arrghh someone at the door. BRB. 1112031234 M * Bertl lol 1112031404 M * SiD3WiNDR ~root, brr :) 1112031433 M * Bertl well, sometimes you need the additional kick ;) 1112031585 J * erwan_ho ~erwan@lns-vlq-39f-81-56-133-136.adsl.proxad.net 1112031591 M * Bertl evening erwan! 1112032125 M * psg OK, I'm back. And armed with another cup of coffee. I ran the testme script, and it fails with chcontext: execvp(grep) Permission Denied. 1112032136 M * maharaja damn 1112032137 M * psg Seems like chcontext is broken. 1112032147 M * maharaja grsec+vs195 is no fun 1112032153 M * maharaja (mergin the rejects) 1112032183 M * Doener psg: sounds like SELinux to me... anything in syslog? 1112032202 M * Bertl psg the output of the testme.sh please? 1112032205 M * psg I'll check, but it's disabled in the kernel config. 1112032211 M * psg ok 1112032251 M * psg Linux-VServer Test [V0.11] (C) 2003-2005 H.Poetzl 1112032251 M * psg chcontext: execvp("grep"): Permission denied 1112032252 M * psg chcontext failed! 1112032252 M * psg New security context is 1 1112032253 M * psg chbind failed! 1112032253 M * psg Linux 2.6.11-rc3-vs1.9.4-rc4drbd x86_64/0.30.205/0.30.205 [Ea] 1112032254 M * psg VCI: 0001:0025 236 1112032305 M * Bertl and now please 'file `which vcontext` 1112032351 M * psg "/usr/local/sbin/vcontext" 1112032365 M * Doener you missed the 'file' ;) 1112032380 M * psg ha! 1112032381 M * psg ok 1112032427 M * psg "/usr/local/sbin/vcontext: ELF 64-bit LSB executable, AMD x86-64, version 1 (SYSV), statically linked, not stripped" 1112032443 M * Bertl okay, that looks fine ... 1112032451 M * Bertl chcontext: execvp("grep"): Permission denied 1112032465 M * Bertl that looks indeed like some strange security kicking in 1112032481 M * Bertl do you get any messages in 'dmesg' or syslog? 1112032525 M * psg looking for SELINUX related ... one moment 1112032589 M * psg Mar 27 23:54:32 vhost1 kernel: SELinux: Disabled at boot. 1112032796 M * psg Right after that there's a log entry "Capability LSM initialized", but I see nothing else that looks germane. 1112032847 M * Hollow Bertl: i read the FDL and reread gentoos policies.. they require the gentoo docs to be licensed under the Creative Commons - Attribution / Share Alike license, though they seem pretty much identical 1112032996 J * Pazzo ~thomas@host130-250.pool8172.interbusiness.it 1112033155 M * Bertl evening Pazzo! 1112033195 M * Bertl Hollow: well, IANAL but sounds pretty compatible ... 1112033222 M * Bertl (unless you want to reduce the GPL) 1112033243 M * Bertl psg: okay, let's try some things 1112033258 M * Pazzo Hi Bertl, hi all! 1112033268 M * psg I'm ready. 1112033275 M * Hollow i will point readers to the original source and tell them the work is licesned under FDL in the appropriate sections, is that ok? 1112033315 M * pusling is there anything like 'vtop' ? 1112033323 M * Hollow pusling: yep, vtop 1112033325 M * Hollow ;) 1112033346 M * Bertl Hollow: well, IMHO that would be limiting the GPDL 1112033380 M * Hollow wtf, another license 1112033397 M * Bertl but I don't think gentoo _insists_ on documentation being in that license (FDL) 1112033412 M * pusling Hollow: oh yeah... I just needed to be root to do it ;) 1112033438 M * Bertl s/FDL/CCA/SAL/ 1112033442 M * Hollow ehm.. gentoo requires it to be creative commons not FDL 1112033477 M * Bertl well, basically that would mean that you cannot use docu for all gnu tools? 1112033510 Q * Pazzo Quit: Download Gaim: http://gaim.sourceforge.net/ 1112033560 M * Hollow i don't know.. maybe i should just kick this part and link to your paper 1112033587 M * Bertl psg: what does chcontext --ctx 1 grep -Ee '[[]|s_context|VxID' /proc/self/status 1112033590 M * Hollow i'm not in the mood of caring about license shit 1112033624 M * psg New security context is 1 1112033624 M * psg chcontext: execvp("grep"): Permission denied 1112033678 M * Bertl you do not by any chance use debian? 1112033685 M * psg No, all FC3 1112033695 M * Bertl k ;) 1112033704 M * Bertl let's try something simpler then: 1112033711 M * Bertl chcontext --ctx 1 true 1112033824 M * psg same thing. "permission denied". Sorry for the delays ... I'm working through a KVM switch that occassionally goes nuts. 1112033832 M * Bertl np 1112033850 M * Bertl so then let's do some basic checks like: 1112033870 M * Bertl ls -lda / /bin /bin/true 1112033906 M * psg drwxr-xr-x 29 root root 4096 Mar 28 10:36 / 1112033907 M * psg drwxr-xr-x 2 root root 4096 Mar 26 04:05 /bin 1112033907 M * psg -rwxr-xr-x 1 root root 18744 Oct 5 11:50 /bin/true 1112033927 M * psg everything else on this box seems to be ok, BTW 1112033932 M * Bertl k, and now ... 1112033948 M * Bertl showattr -d / /bin /bin/true 1112033994 M * psg no ouptu from that whatsoever ... just another prompt. 1112034003 M * psg output 1112034024 M * Bertl interesting ... 1112034027 M * Hollow gtg, later guys! 1112034032 M * psg bye 1112034033 M * Bertl cya Hollow! 1112034116 M * psg Could this be a problem with dietlibc ? 1112034149 M * Bertl psg: okay please uplaod the output of 'vserver-info - SYSINFO' somewhere (e.g. pastebin.com) 1112034224 M * psg Not much there, just: Kernel: 2.6.11-rc3-vs1.9.4-rc4drbd 1112034224 M * psg VS-API: 1112034231 M * psg and that's it. 1112034253 M * Bertl well, somehow your tools got broken ... 1112034264 M * Bertl let's try to strace that 1112034278 M * mikegrb Bertl: <3 1112034303 M * Bertl 'strace -fF 1112034306 M * psg They sure did. OK ... you'll have to be real explicit ... not real familiar with the debugging tools. 1112034325 M * Bertl strace -fF -o some.trace vserver-info - SYSINFO 1112034332 M * psg thx 1112034345 M * Bertl hey mikegrb! ;) 1112034382 M * psg 11598 execve("/usr/local/sbin/vserver-info", ["vserver-info", "-", "SYSINFO"], [/* 39 vars */]) = 0 1112034383 M * psg 11598 open("/dev/urandom", O_RDONLY) = 3 1112034383 M * psg 11598 read(3, "\3538", 2) = 2 1112034384 M * psg 11598 read(3, ":\23@\363\2\224\34\334", 8) = 8 1112034384 M * psg 11598 close(3) = 0 1112034384 M * psg 11598 open("/usr/local/lib/util-vserver/FEATURES.txt", O_RDONLY|0x8000) = 3 1112034385 M * psg 11598 uname({sys="Linux", node="vhost1.latpro.com", ...}) = 0 1112034386 M * psg 11598 write(1, "Versions:\n Ker"..., 37) = 37 1112034386 M * psg 11598 write(1, "2.6.11-rc3-vs1.9.4-rc4drbd", 26) = 26 1112034388 M * psg 11598 write(1, "\n VS-API: ", 28) = 28 1112034388 M * psg 11598 _exit(236) = ? 1112034457 M * Bertl well, that means that the build is basically broken ... 1112034472 M * Bertl otherwise the file /usr/local/lib/util-vserver/FEATURES.txt would contain the info 1112034497 M * Bertl how did you build the tools? which compiler and libs? 1112034512 M * psg OK, should I rebuild without dietlibc this time? I know that the previous version of dietlibc had build problems with AMD64. 1112034528 M * psg gcc ... everything was stock FC3 install. 1112034532 M * Bertl try that, or get a working dietlibc ;) 1112034555 M * psg I'll try it and come back when I'm done. 5 minutes or so. 1112034560 M * erwan_ho evening Bertl 1112034653 M * maharaja whats the best way to test if my vs195-grsec merge suceeded? 1112034741 M * Bertl use it with some vservers ;) 1112034758 M * maharaja well - until now i used: 2.4.26-vs1.3.9-xfsacl-yule 1112034777 M * Bertl guess it can only get better then .. *G* 1112034787 M * maharaja now i try to do grsec-vs1.9.5 1112034802 M * maharaja the "problem" is - i will not use it at the same system :) 1112034815 M * maharaja never touch a running system ;) 1112034830 M * maharaja i'm setting up a new system not knowing what i do *g* 1112034836 M * maharaja (production system) 1112034839 M * maharaja *dumidudm* 1112034905 M * Bertl well, the testme.sh is for sure a good test 1112034913 M * Bertl (use it with -L option too) 1112034916 M * psg Bertl: That was it. Without dietlibc it works (at least vserver-stat ... haven't tried everything.) I thought that I had already tested that, but obviously I had not done what I thought I'd done. 1112034925 M * psg Thanks so much for your help. 1112034955 M * Bertl you're welcome, so dietlibc (which version?) is broken on x86_64/fc3? 1112034960 M * psg Now I'm off to contem,plate the beauties of "Vhashify" !!! 1112034969 M * psg The latest ---- 0.28 1112034979 M * maharaja *searching for testme.sh* 1112034985 M * psg 0.27 would not build, 0.28 builds, but doesn't work. 1112035136 M * Bertl k, thanks! 1112035274 P * psg 1112035785 Q * lilo Quit: bbiab 1112036143 J * lilo ~lilo@lilo.usercloak.oftc.net 1112036345 M * Bertl wb lilo!! 1112036364 J * Borg_Number_One ~borg_numb@p54B89149.dip0.t-ipconnect.de 1112036395 M * Borg_Number_One hi 1112036405 M * Bertl hey B#1, looking for sladen? 1112036413 M * Borg_Number_One yes. 1112036609 M * Borg_Number_One Well, sladen seems to be offline. 1112036651 M * Bertl we'll let him know that you were looking for him ... 1112036748 P * Radiance - 1112036806 A * sladen hmmms from Amsterdam 1112036850 M * Borg_Number_One Can someone build screenshots from Phoenix-BIOS-Editor while it runs in Linux's Wine or Win4Lin ? 1112036953 M * Bertl no wine or win4lin here ... 1112037442 M * Borg_Number_One cu 1112037445 P * Borg_Number_One 1112040872 Q * lilo Quit: brb 1112041434 J * lilo ~lilo@lilo.usercloak.oftc.net 1112043231 Q * lilo Quit: brb 1112045032 Q * grecea Ping timeout: 480 seconds 1112045960 J * tad ~truex@static-151-204-232-50.bos.east.verizon.net 1112046002 M * Bertl welcome tad! 1112046015 M * tad Hi Bertl (et. al.) 1112046525 J * Martin ~mjb@sev.gotadsl.co.uk 1112046536 M * Bertl welcome Martin! 1112046540 M * Martin Hi 1112046563 M * Martin Was wondering if I could ask about non-x86 platforms 1112046578 M * Martin As in - does vserver work on them? 1112046597 M * Bertl sure .. 1112046602 J * Psy0rz psy0rz@linuxsysteem.nl 1112046607 M * Martin Cool 1112046616 M * Bertl http://linux-vserver.org/Syscall+Switch+Info 1112046622 M * Bertl welcome Psy0rz! 1112046623 M * eyck yeah, we even got official vserver-syscall for ppc 1112046625 M * erwan_ho Martin: what platform are yoy thinking about ? 1112046636 M * eyck probably arm7tdmi 1112046637 M * Psy0rz thx Bertl 1112046649 M * Martin erwan_ho: SMP sun4u - and E4000 to be precise 1112046663 M * Bertl so sparc64 then ... right? 1112046672 M * Martin Yep 1112046679 M * eyck nice 1112046700 M * Bertl Martin: it's not tested yet, but it should work fine, or at least can be made working with minimal efford 1112046721 M * Bertl we had no sparc64 recently available ... 1112046721 M * Martin OK 1112046734 M * Martin Any idea what effort is likely to be needed? 1112046739 M * Psy0rz 'Can't set ipv4 root' 1112046745 M * Psy0rz probably a common problem :) 1112046768 M * Bertl Psy0rz: soem details? maybe output from testme.sh? 1112046785 M * Bertl Martin: would say an hour or two on the irc channel ;) 1112046807 M * Martin Bertl: That I can do - before or after the compile :-D 1112046845 M * Martin What non-x86 platforms are needed for testing / would be useful? 1112046853 M * Bertl after the compile ;) 1112046863 M * Martin Bertl: :-) 1112046869 M * eyck hmm, I think only PPC is well tested 1112046887 M * Bertl well, we tested x86 and x86_64 yet there was some testing on sparc, and hppa 1112046909 M * eyck hppa? when? 1112046914 M * Bertl eyck: who tested on ppc? 1112046923 M * Bertl eyck: some time ago ... 1112046935 M * Martin I've recently cut down on the number of architectures I own but if it's useful I can try to get some testing on a few of them 1112046945 M * eyck Bertl: I don't recall name, but I remember some guy checking out ppc 1112046956 M * Bertl Martin: that would be _really_ great! 1112046983 M * eyck Bertl: oh well, I'll try sitting silently in a corner for a change now. 1112046992 M * Bertl Martin: for sure would gain you a place in our Hall o' Fame 1112047010 M * Martin Bertl: OK I'll see what can be done 1112047030 M * Martin Can't promise it to be regular but a few one off testing runs might be possible. 1112047057 M * Bertl well, I'd say an initial testing on non x86 arch, until it works once, is sufficient 1112047068 M * Bertl the arch specific code is minimal 1112047096 M * eyck oh, and qemu supports emulating ppc, 1112047106 M * Martin That's cool, I'll have a word around some of the multi-arch collectors I know 1112047114 M * Martin qemu - that's a point. 1112047120 M * Bertl that would be even better! 1112047122 M * eyck hmm, I even fail at sitting quietly. 1112047141 M * Martin Think they have alpha support for some other platforms (i/c alpha) 1112047148 M * eyck even at sitting quietly I fail miserably? 1112047154 M * eyck much better now, thank you aziz. 1112047174 M * Bertl eyck: yeah, but we love you anyways ;) 1112047181 M * eyck Martin: haven't seen working alpha on qemu? 1112047194 M * eyck yeah, I know I'm ugly, but my mama loves me. 1112047209 M * Martin eyck: I meant alpha in terms of development rather than architecture 1112047222 M * Bertl yeah, we use that term too often too ;) 1112047231 M * eyck midnight. time for lowly creatures to crawl back to their lairs 1112047291 M * Psy0rz Bertl 1112047292 M * Psy0rz root@banaan util-vserver-0.30 # vserver stage3apache start 1112047292 M * Psy0rz Starting the virtual server stage3apache 1112047292 M * Psy0rz Server stage3apache is not running 1112047292 M * Psy0rz Can't set the ipv4 root (Function not implemented) 1112047302 M * Psy0rz couldn't find a test.sh btw 1112047322 M * eyck Psy0rz: what is your kernel's config? 1112047330 M * Psy0rz looked like no kernel support to me 1112047331 M * Psy0rz hold on 1112047381 M * Martin eyck: http://fabrice.bellard.free.fr/qemu/status.html - support for emulating ARM, SPARC and SPARC64 in the works 1112047400 M * Martin Out of interest what platform specific code is there and what does it do? 1112047440 M * Psy0rz #CONFIG_VSERVER_LEGACY=y 1112047440 M * Psy0rz CONFIG_VSERVER_PROC_SECURE=y 1112047440 M * Psy0rz CONFIG_VSERVER_HARDCPU=y 1112047448 M * Psy0rz and UGID24=y 1112047477 M * Bertl it's the syscall and some copies of the utsname stuff which is unfortunately platform specific 1112047497 M * Psy0rz oh that sucks 1112047500 M * Psy0rz i'm using gentoo 1112047508 M * Martin Cool - not a lot 1112047509 M * Bertl Psy0rz: that's 2.6 kernel and an older aptch, right? 1112047526 M * Psy0rz hold on 1112047541 M * Bertl Martin: no, we have some kind of multiplexor syscall with fixed size (C99) types 1112047553 M * Bertl (types in structures, that is) 1112047564 M * Martin OK 1112047571 M * eyck Martin: yeah, I just haven't seen anything working yet. 1112047591 M * Bertl Martin: qemu i386 on i386 and ppc works fine ... 1112047597 M * eyck besides, Bertl is our qemu liaison 1112047602 M * Bertl (we use it for development and testing) 1112047612 M * tad quit 1112047616 Q * tad Quit: using sirc version 2.211+KSIRC/1.3.10 1112047633 M * Martin I don't claim to be an expert by any means 1112047636 M * Bertl ppc emulation on x86 was not working last time I checked (system emulation that is) 1112047639 M * Psy0rz Linux banaan 2.6.11-rc3-vs1.9.4 Bertl 1112047652 M * eyck hmm, last time I checked == cvs? 1112047652 M * Martin Good to know what does and doesn't work on qemu though. 1112047654 M * Martin Ta. 1112047664 M * eyck 0.6.1 works fine. 1112047671 M * Psy0rz 1.9.4 is old? 1112047672 M * Bertl Psy0rz: okay, upgrading to 1.9.5 would help, but enabling the legacy mode would do the trick too 1112047683 M * Bertl CONFIG_VSERVER_LEGACY=y 1112047686 M * Psy0rz well it is enabled (ignore that #) 1112047691 M * eyck Psy0rz: in normal or internet years? 1112047694 M * Psy0rz copy/pasting in screen is a bitch :D 1112047705 M * Bertl hmm, is it? 1112047710 M * Martin Anyway, gotta go - hopefully I shall return with reports of working systems :-D 1112047716 M * Psy0rz when you're used to kclipper and kde it is 1112047725 M * Bertl Martin: okay, thanks in advance, and cya! 1112047729 M * eyck hmm, copy/pasting in screen rocks. 1112047732 Q * Martin Quit: using sirc version 2.211+KSIRC/1.1 1112047740 M * Psy0rz :) 1112047747 M * Psy0rz screen rocks period ;) 1112047752 M * Bertl Psy0rz: okay, let's try the testme.sh 1112047756 M * Psy0rz k 1112047759 M * Psy0rz whree to get it? 1112047766 M * eyck from internet. 1112047768 M * Psy0rz lol 1112047770 M * Bertl http://vserver.13thfloor.at/Stuff/SCRIPT/testme.sh 1112047770 M * Psy0rz google 1112047771 M * Psy0rz :) 1112047774 M * Psy0rz thx 1112047792 M * eyck K, enough is enough, if you need some background offtopic activity, just call me. nighty night. 1112047844 M * Psy0rz lol 1112047848 M * Psy0rz we will 1112047849 M * Psy0rz :) 1112047855 M * Psy0rz whats your number? 1112047861 M * Psy0rz Bertl can i just copy the quicktest output? 1112047873 M * eyck go ahead, make my day 1112047877 M * Bertl well, msg me in private 1112047880 M * Psy0rz k 1112047896 M * eyck no more pastebin? 1112047926 M * Bertl eyck: too much typing ;) 1112047942 M * eyck uff... I always hated pastebin 1112047945 M * Bertl Psy0rz: are you sure that legacy is enabled in your kernel? 1112047954 M * Psy0rz pastebin is pretty annoying i think :) 1112047962 M * Psy0rz Bertl not anymore 1112047963 M * Psy0rz i'll check again 1112047966 M * eyck damn right it's annoying. 1112047982 M * Psy0rz probably forgot to intall the bzimage or something lame 1112047996 M * Bertl ah, yes, that sounds even more like it 1112047997 M * Psy0rz why is it called legacy support anyway, when the tools arent ready yet? :) 1112048009 M * daniel_hozac eyck: and people flooding channels isn't? :) 1112048011 M * Psy0rz it should be called future-legacy-support ;) 1112048021 M * Bertl you are using the legacy tools, what do you expect ;) 1112048034 M * Psy0rz where to get the 1337 new tools? :) 1112048037 M * eyck daniel_hozac: nope. 1112048040 M * Psy0rz cvs? 1112048052 M * Bertl nope, just get util-vserver 0.30.205 1112048058 M * Psy0rz ah 1112048059 M * Bertl (it's the alpha branch) 1112048065 M * Psy0rz didn't compile in gentoo yet :( 1112048069 M * Psy0rz alpha 1112048070 M * Psy0rz ok 1112048075 M * Bertl then you can disable legacy support in 1.9.5++ 1112048082 M * Psy0rz i'll stick with good old legacy then 1112048143 M * Bertl eyck: let me paste my new patch to you in private ;) 1112048163 M * Bertl (it's just a few hundred lines long ;) 1112048214 M * Psy0rz lol 1112048219 M * Psy0rz gzip it and then patch it uuencoded ;) 1112048263 M * Psy0rz s/patch/paste 1112048265 M * Psy0rz it's late here 1112048273 M * Psy0rz maybe i should logout as root now :) 1112048280 M * Bertl well, it's early here ;) 1112048303 M * Psy0rz 00:18 1112048306 M * Psy0rz actually not htat late 1112048310 M * Psy0rz but i'm sleepy 1112048313 M * Bertl yep, same here ;) 1112048326 M * daniel_hozac middle of the day :) 1112048337 M * Bertl right after noon ... ;) 1112048500 M * erwan_ho night all 1112048513 Q * erwan_ho Quit: Teleporting To Bed : [Ok] 1112049078 M * Psy0rz i'll try again tomorrow :) 1112049081 M * Psy0rz nightynight 1112049100 M * Bertl k, good night then! 1112050451 J * lilo ~lilo@lilo.usercloak.oftc.net 1112051057 Q * lilo Quit: oops, forgot two PCI cards 1112051406 M * Bertl ah, seems like my pgalloc cleanup makes some progress ... 1112051445 M * Bertl (was signed off by akpm and forwarded to linus) 1112051875 Q * ruut Quit: Nettalk6 der Freeware IRC-Client 1112051915 J * DuckMaster ~Duck@dyn-83-157-187-36.ppp.tiscali.fr 1112051999 Q * yarihm Quit: Leaving 1112052332 Q * DuckKing Ping timeout: 480 seconds 1112052613 M * Bertl okay folks, I'm a little tired, so I'm off to bed now .. 1112052646 M * Bertl have a nice whatever everyone, cya tomorrow ... 1112052674 N * Bertl Bertl_zZ 1112053914 J * ciphernaut ~a@61.88.18.130