1104797675 Q * we2by Read error: Connection reset by peer 1104797683 J * we2by ~we2by@dc5146d009.adsl.wanadoo.nl 1104797856 J * shuri ~shuri@dsl.speedline209.227.electronicbox.net 1104799134 M * we2by vserver-stat does not give a list of all vservers 1104799661 M * sannes :) 1104799756 M * we2by I found out 1104800180 Q * flock Ping timeout: 480 seconds 1104800626 J * flock ~restless@l192-117-111-12.broadband.actcom.net.il 1104800904 Q * shuri Remote host closed the connection 1104805215 J * matti matti@linux.gentoo.pl 1104806156 J * shuri ~shuri@dsl.speedline209.227.electronicbox.net 1104806750 N * Bertl_zZ Bertl 1104813257 Q * shuri Quit: Leaving 1104813919 J * nox- ~nox@213.39.207.36 1104814246 Q * nox Ping timeout: 480 seconds 1104814278 N * nox- nox 1104814475 Q * sannes Read error: Connection reset by peer 1104815145 Q * mikeb Quit: Leaving 1104815483 Q * click Ping timeout: 480 seconds 1104817060 Q * we2by Ping timeout: 480 seconds 1104817083 J * we2by ~we2by@dc5146d009.adsl.wanadoo.nl 1104818487 Q * _are_ Quit: Disconnecting 1104819708 Q * infowolfe Read error: Connection reset by peer 1104821204 J * infowolfe ~infowolfe@209-112-186-118-cdsl-rb1.nwc.acsalaska.net 1104821458 J * sannes ~ace@home.skarby.no 1104822062 M * Bertl k, back later ... 1104822068 N * Bertl Bertl_oO 1104822086 J * _are_ ~are@gateway-dsl.lihas.de 1104822177 M * _are_ hi 1104822921 M * infowolfe keanu: server status? 1104822927 M * infowolfe bunny is asking 1104823070 N * Bertl_oO Bertl 1104823081 M * Bertl morning folks! 1104823152 M * Bertl infowolfe: http://vserver.13thfloor.at/Experimental/NGNET/diff-2.6.10-vs1.9.3.16-ng8.7.diff 1104823247 M * infowolfe Bertl: thanks, but i was going to apply that to keanu's server, before he told bonsaikitten that he'd rather do the kernel himself 1104823280 M * Bertl well, that's fine with me ;) 1104823289 Q * infowolfe Quit: (changing kernels) 1104823736 J * infowolfe ~infowolfe@209-112-217-248-cdsl-rb1.nwc.acsalaska.net 1104824437 M * _are_ server up again with 2.6.10-vs1.9.3.16 and no NETPOLL stuff. 1104824466 M * _are_ now doing some network load and hope for a stable server 1104824675 M * Bertl k, keep me updated 1104824899 M * Loki|muh we2by: what was it? 1104827416 M * Bertl k, back later ... 1104827421 N * Bertl Bertl_oO 1104829470 Q * sannes Read error: Connection reset by peer 1104831539 M * we2by Loki|muh, two vserver have the same context value 1104831588 M * Loki|muh ah 1104832205 Q * infowolfe Read error: Connection reset by peer 1104832264 Q * albeiro Remote host closed the connection 1104832268 J * albeiro albeiro@linux.gentoo.pl 1104832283 J * infowolfe ~infowolfe@209-112-213-97-cdsl-rb1.nwc.acsalaska.net 1104832285 Q * infowolfe Quit: 1104832380 J * infowolfe ~infowolfe@209-112-213-97-cdsl-rb1.nwc.acsalaska.net 1104832551 N * Bertl_oO Bertl 1104832576 M * Bertl morning again! everything fine? 1104832587 M * infowolfe yay, xchatosd rocks ;-) 1104832676 M * Bertl hehe, what about keanu and his kernels? 1104834429 M * Doener morning! 1104834471 M * we2by morning! 1104834475 M * we2by :-) 1104834504 M * we2by I have 4 vservers running :--) 1104834509 A * Bertl .o( another day in the year of the Linux-Vserver ;) 1104834521 M * Bertl we2by: congrats! 1104834524 M * Doener *g* 1104834539 M * we2by but something weird happen when I sleep 1104834551 M * we2by all my putties get disconnected when I woke up 1104834603 M * Loki|muh maybe you are a sleepwalker? ;) 1104834615 M * we2by nope 1104834634 M * Doener 06:37:40 * we2by has quit (Ping timeout: 480 seconds) 1104834635 M * Doener 06:38:03 * we2by (~we2by@dc5146d009.adsl.wanadoo.nl) has joined #vserver 1104834644 M * Doener looks like your connection was interrupted 1104834648 M * we2by maybe its my damn router 1104835289 M * ndim Hi. /aw 1104835322 M * ndim I'm about to build myself a new kernel with Linus' ACPI Video fix. Should I ad ngnet or not? 1104835346 M * Bertl if you like, but don't expect it to work out of the box ;) 1104835402 M * Doener Bertl: what name did we choose for a patch that 'adds only the last'? 1104835444 M * Doener ah, diff 1104835499 M * Doener Bertl: http://doener.homeip.net/doener/vserver/diff-2.6.10-vs1.9.3.16-openfd.diff 1104835513 M * Doener still needs some cleanup 1104835536 M * Bertl the important question is: does it work? 1104835580 M * Doener Jan 03 07:48:35 Doener Bertl_zZ: hm, openfd accounting works for me. IIRC didn't change anything significant compared to the openfd code in 1.9.2.3 ... 1104835580 M * Doener Jan 03 07:48:57 Doener f.e. this seems to work just fine: ls | chcontext --ctx 100 ls `chcontext --ctx 200 ls` 1104835580 M * Doener Jan 03 07:49:17 Doener both contexts never go below zero and reach zero at the end 1104835627 M * Doener tests done with qemu, results are based on the output of the debug stuff 1104835667 M * Bertl oaky, I'll test that ... 1104835691 M * Doener want my qemu test kernel? 1104835865 M * Doener http://doener.homeip.net/doener/vserver/kernels/qemu/ 1104836140 M * Bertl no, thanks, I compile my own :P 1104836180 M * Doener ok, fine 1104836439 J * sannes ~ace@home.skarby.no 1104836455 M * Bertl welcome sannes! 1104837813 M * Bertl Doener: let's make the VLIMIT_OPENFD and use 17 for that ... 1104837820 M * Bertl s/the/that/ 1104837830 M * Doener so it works? 1104837848 M * Bertl had not time to test it yet ... just reading the patch ;) 1104837853 M * Doener ah ok 1104837890 M * Doener what about the fd? shall i just cast it to a pointer? 1104837956 M * Doener ... in vx_openfd_(inc|dec) 1104837958 J * CounteR NemanjaTx@81.93.88.47 1104837963 M * meebey hiya 1104837984 M * meebey which /dev/* do I need inside vservers? I wanna close _my_ security hole :) 1104837997 T * CounteR Nemanja VIII2 i Nikolina VIII4 1104838026 T * Doener http://linux-vserver.org/ | latest stable 1.29, devel 1.3.9, 1.9.3, ng8.7 1104838048 M * Bertl CounteR: hmm? 1104838050 M * meebey wtf 1104838066 M * Bertl $ ls /dev/ 1104838066 M * Bertl full log= null ptmx pts/ random tty urandom zero 1104838070 M * CounteR What 1104838118 M * meebey Bertl: full= 1104838122 M * meebey Bertl: uh? 1104838138 M * meebey opposite of null? 1104838151 M * Bertl yeah, but not required ... 1104838172 M * Bertl but you might want console or tty (depends on the usage) 1104838185 M * meebey I only log into those via vserver foo enter 1104838188 M * meebey or sshd 1104838193 M * sannes Trid to use 1.9.3.16 now .. and I get segmentation faults.. 1104838206 M * sannes anyone experienced the same? 1104838217 M * Bertl interesting, yeah, seems like keanu got the same issues ... 1104838256 M * sannes and I got something really nice in the log if you want it.. 1104838290 F * ChanServ +o Doener 1104838302 M * Doener hu? 1104838314 M * Bertl okay, I'm off for lunch now ... back later ;) 1104838328 N * Bertl Bertl_oO 1104838343 F * Doener +t 1104838358 M * Doener same here ;) 1104838361 N * Doener Doener|gone 1104838408 M * sannes what was it I were to run it thorugh again.. ksymoops ? 1104839009 N * Doener|gone Doener 1104839207 F * Doener -t 1104839373 M * meebey Bertl_oO: ping 1104839385 Q * CounteR Read error: No route to host 1104839388 M * meebey I got a feature request/idea for vserver (more security) 1104839412 Q * Loki|muh Quit: bbl 1104839429 M * Doener meebey: just tell about it ;) 1104839456 M * meebey its about the CAP_SYS_IO 1104839466 M * meebey it seems like its by default allowed 1104839473 M * meebey it should be not required 1104839506 M * meebey brb 1104839679 M * meebey re 1104839712 M * meebey ups I was wrong 1104839719 M * meebey CAP_SYS_IO is not alled by default 1104839732 M * meebey nevermind, vserver rocks! ;) 1104839799 M * Doener you mean CAP_SYS_RAWIO ? 1104839812 M * meebey not sure, the one that allows you blockdevice access 1104839814 M * meebey like /dev/hda1 1104840045 J * Loki|muh loki@satanix.de 1104840257 M * meebey aaargs 1104840268 M * meebey the guy who made woody images didn't create a directory! damnit 1104840269 M * Doener hm? 1104840338 M * meebey now I got ./dev and ./etc and so on 1104840379 M * Loki|muh hehe 1104840444 M * meebey instead of woody_something/dev... 1104840446 N * Bertl_oO Bertl 1104840454 M * sannes .. scheduling while atomic: chbind, what does that actually mean? If it wern't supposed to be interrupted while doing whatever it was doing, why was interrupts enabled? 1104840459 M * meebey probably a non-unix/linux guy 1104840487 M * meebey tar -cjvf foo.tar.bz2 foo/ is better than tar -cjvf foo.tar.bz2 foo/* 1104840502 M * meebey and when you extract then its better too 1104840507 M * Bertl sannes: do you have preemption enabled? 1104840509 M * meebey no need to create a directory... 1104840520 M * meebey gone, lunch 1104840536 M * sannes Bertl : yes, compiling it now without it.. felt brave after putting in a couple of fixes for 2.6.. heh 1104840578 M * Bertl do you ahve the oops/trace/log? 1104840588 M * sannes yip 1104840603 M * sannes but I wasn't able to run ksymoops on it.. 1104840647 M * sannes it gives errors.. the more options I give it, the more errors.. heh 1104840749 M * sannes http://www.sannes.org/panics/segmessages 1104840869 M * Bertl hmm, do you get one message everytime the chbind is called? 1104840944 M * sannes yes, . unfortunatly i ran it a couple of times before I relized that there was output in the log.. 1104840974 M * Bertl could you try something on that machine for me? 1104841027 M * sannes sure, .. and I get segmentation fault if i run it just on the cmd... 1104841063 M * Bertl try 'chbind --ip 127.0.0.1 -- true' 1104841079 M * Bertl mark the log before that .. so that we see just the _new_ output 1104841343 M * Bertl Doener: looks to me like 'something' in 1.9.3.16 makes chbind segfault under certain conditions ... 1104841375 M * Bertl the scheduling while atomic (sannes observes) IMHO is just a side effect of the core dump ... 1104841486 M * sannes http://www.sannes.org/panics/segmessages2 1104841546 M * Bertl okay, could you try an strace -fF on the chbind? 1104841712 M * sannes http://www.sannes.org/panics/chbind-strace 1104841717 M * Bertl tx 1104841740 M * sannes heh, fancy, want me to boot on a non-preemptive ... ? 1104841749 M * Bertl you are not using gentoo, by any chance? 1104841762 M * sannes matter of fact I am.. 1104841780 M * Bertl interesting ... 1104841786 M * sannes but it worked before I changed to 1.9.3.16 .. (from 1.9.3.14) 1104841813 M * Bertl yeah, exactly the same thing was reported by keanu/infowolfe yesterday ... 1104841844 M * Bertl atm, I have no clue what it is, but I actually suspect the kernel ... 1104841855 M * Bertl (or to precise my latest changes ;) 1104841856 M * sannes maybe a bad merge? 1104841870 M * sannes if the patch utility has been changed .. heh 1104841876 M * sannes would have been fun.. hehehe 1104841910 M * Bertl could you compile a new 2.6.10-vs1.9.3.14 with the same settings you have for the .16 for me and test that one? 1104841933 M * Bertl (without changing anything else, not even the preempt stuff) 1104841971 M * sannes that is going to be a bit of a problem, but I copied the .config raw over from that so it should be the same.. 1104842060 M * sannes with the exception of preempt that is.. 1104842094 M * sannes but I know for a fact that it worked with preempt on (excpet that the kernel crashed with and without it after a while) 1104842137 M * sannes going to patch up a 2.6.10-vs1.9.3.14 now .. going to take a little while 1104842205 M * Bertl np, thanks! 1104842469 M * sannes ok, compiling it now ;) 1104842638 J * JonB ~NoSuchUse@cpe.atm2-0-1031198.0x50a4ad0e.bynxx13.customer.tele.dk 1104842644 M * JonB hey bertl 1104842645 M * Bertl welcome JonB! 1104842658 M * JonB Bertl: http://vserver.13thfloor.at/Remote-Reset/ 1104842667 M * JonB Bertl: i need this for 2 machines at work 1104842678 M * JonB i suppose i should use data0 and 1 for machine a 1104842684 M * JonB and data 2 and 3 for machine b ? 1104842696 M * Bertl for example ... 1104842708 M * JonB and then in the parport util type parport -t 4 1 1104842709 M * JonB 0 1104842713 M * JonB or 2 ? 1104842720 M * JonB -t for 4 seconds for ATX 1104842759 M * Bertl well, would be 1/2 or 4/8 depending on your wiring 1104842769 M * JonB what do you mean ? 1104842790 M * Bertl 0,1 (data port) -> 1,2 decimal 1104842802 M * Bertl 2,3 (data port) -> 4,8 decimal 1104842835 M * JonB well, at home i used data 0 and data1 1104842854 M * JonB and the ssh authorized keys contains -t 2 1 1104842878 M * Bertl yep , which means to toggle the data=0 (1) line 1104842890 M * JonB oh 1104842909 M * JonB so i need a parport -t 4 -p 1 1 1104842912 M * JonB so i need a parport -t 4 -p 0 1 1104842918 M * JonB so i need a parport -t 4 -p 2 1 1104842919 M * JonB so i need a parport -t 4 -p 3 1 1104842920 M * Bertl (at least IIRC, check the source ;) 1104842921 M * JonB one of 1104842931 M * JonB now i see 1104842945 M * Bertl you probably need parport -t 4 4 1104842948 M * JonB depending on if i put the input on data0/data1 1104842985 M * Bertl (now I got you confused I guess ;) 1104843000 M * JonB Bertl:This is ./parport V2.4 1104843000 M * JonB options are: 1104843000 M * JonB -h print this help message 1104843000 M * JonB -p select parport id (0 default) 1104843012 M * JonB where is the -p used? 1104843021 M * JonB if i use a different parallel port? 1104843028 M * Bertl yep, exactly 1104843034 M * JonB oh 1104843037 M * Bertl you can have up to 3 parports 1104843072 M * JonB i thought it was the line 1104843073 M * JonB okay 1104843088 Q * keanu Quit: leaving 1104843112 M * JonB hmm i think i'll manage with just one controller machines, i dont have that many servers here at work 1104843131 M * JonB how long a cable do you think i can use ? 1104843160 M * JonB i thought of connecting the slave machines using phone patch cables 1104843164 M * sannes Bertl : booted and works like a charm.. 1104843179 M * sannes exception is that the utils are complaining about something.. 1104843185 M * sannes some tmp files.. 1104843187 M * sannes sec 1104843199 M * Bertl okay, great! let me examine the changes we did between .14 and .16 1104843202 M * sannes chbind atleast works :) 1104843224 J * nox- ~nox@213.39.207.36 1104843361 Q * nox Ping timeout: 480 seconds 1104843390 M * JonB Bertl: thanks, later 1104843391 Q * JonB Quit: Leaving 1104843404 N * nox- nox 1104843556 J * JonB ~NoSuchUse@cpe.atm2-0-1031198.0x50a4ad0e.bynxx13.customer.tele.dk 1104843574 M * Bertl wb JonB! ;) 1104843580 M * JonB Bertl: regularly null-modem cable, right? 1104843583 M * sannes it all seems to work fine now :) what changes am I missing sticking to 1.9.3.14 ? 1104843610 M * Bertl a few supposed-to-be fixes 1104843629 M * Bertl JonB: for what? 1104843637 M * JonB Bertl: console on serial port 1104843663 M * Bertl yep, pinout and wiring is on my pages ... 1104843674 M * JonB Bertl: thanks 1104843816 M * Bertl sannes: it would be good to identify and fix this issue, if this isn't a problem for you? 1104843836 M * sannes Bertl : I want it fixed :) just tell me what you need... 1104843914 M * Bertl k, great! do you have CONFIG_VSERVER_DEBUG enabled? 1104844043 Q * JonB Quit: Leaving 1104844077 M * sannes nopes, should I enable it_ 1104844081 M * sannes ? 1104844089 M * Bertl yep, would probably help 1104844096 M * sannes on both? 1104844175 M * Bertl sec, I guess I found it ;) 1104844195 M * Bertl you have the 1.9.3.16 source tree at hand? 1104844261 M * sannes yip 1104844271 M * Doener ah, __nx_dynamic_id, right? 1104844278 M * Bertl nope ... 1104844286 M * Doener out_unlock: 1104844286 M * Doener spin_unlock(&nx_info_hash_lock); 1104844286 M * Doener + rcu_read_lock(); 1104844293 M * Doener i would say so... 1104844296 M * Bertl yep 1104844323 M * Bertl sannes: okay, change line 249 1104844330 M * Bertl of kernel/vserver/network.c 1104844347 M * Bertl from rcu_read_lock(); to rcu_read_unlock(); 1104844373 M * Bertl and reenable the preempt stuff (if disabled) 1104844430 M * Doener Bertl: what did you have in mind? 1104844692 M * Bertl I was just confused by __nx_dynamic_id, because it's in __loc_nx_info() ;) 1104844700 M * Doener ah ok :-) 1104844729 M * Doener evil inlining! *g* 1104844737 M * sannes Bertl : might need a little help with vroot .. get operation not permitted when trying to enable quota.. 1104844750 M * Doener CAP_QUOTACTL ? 1104844754 M * sannes ah 1104844755 M * sannes heh 1104844756 M * sannes nevermind 1104844758 M * Doener ;) 1104844759 M * sannes :> 1104844763 M * Bertl LOL 1104844777 M * sannes I knew I forgot to convert something when changing over.. 1104844797 M * sannes no where on earth do I enable it.. hehe 1104844840 M * Bertl Doener: what makes me really wonder is, why does this bug not hit on the quad PIII? 1104844963 M * Bertl Doener: hmm, do we have a CAP_QUOTACTL yet in 2.6? ;) 1104845011 M * Doener dunno ;) just checked the flowerpage/alpha tools... they define it for 2.6 kernels 1104845209 M * meebey Bertl: is initctl in a vserver required/useful? 1104845231 M * Doener meebey: if you want to control the host, yes ;) 1104845236 M * Bertl sannes: let me check what cap you need (and if we need to add CAP_QUOTACTL) 1104845254 M * Bertl meebey: makes sense if you have a real init inside your vserver 1104845256 M * sannes ok, rebooted with the patch applied and everything seems good! :) 1104845267 M * meebey Bertl: nope I dont 1104845282 M * Bertl then it's not extremely useful ;) 1104845284 M * meebey so no initctl, good 1104845325 M * sannes If I put initctl into a vserver is it virtualized? 1104845363 M * Doener Bertl: my best answer for PIII is: no CONFIG_PREEMPT, or not using network contexts... 1104845376 M * Bertl sannes: it should be ... 1104845379 M * meebey Doener: no preempt? 1104845395 M * Bertl Doener: I guess the preempt is the key here, I avoid it like hell ;) 1104845419 M * sannes heh, or extreme "luck" :> 1104845434 M * Doener meebey: rcu_read_lock() is defined as preempt_disable() 1104845442 M * meebey uh ic 1104845449 M * Bertl no, the quad is running stress tests so no chance for luck ;) 1104845467 M * sannes hehe, you never know... hehe 1104845468 M * meebey so 2.4 there is no preempt so problem there :-P 1104845476 M * meebey s/so/with/ 1104845483 M * Doener meebey: only 2.6.10-vs1.9.3.16 contains that bug... 1104845514 M * Doener i was answering: 14:20:40 Bertl Doener: what makes me really wonder is, why does this bug not hit on the quad PIII? 1104845517 M * sannes that explains why it was scheduling while preempt was off.. heh 1104845527 M * Doener s/was/were/ 1104845559 M * Bertl sannes: for a test you could give CAP_SYS_ADMIN to the vserver (for vroot/quota) but I guess I'll add the CAP_QUOTACTL to it ;) 1104845593 M * Doener smells like cap masking ;) 1104846008 M * Bertl sannes: okay, I'll update the vroot patch for 1.9.3.16, is that fine for you? 1104846029 M * Bertl (including the CAP_QUOTACTL) 1104846038 M * sannes ok, Ive obviously misunderstood something.. /etc/vservers/ugh/capabilities right? 1104846084 M * sannes Bertl : fine by me, vroot is always the last patch I add.. 1104846088 M * Bertl yep, that looks good ... 1104846116 M * sannes how can I check that CAP_SYS_ADMIN really was sat? 1104846145 M * Bertl grep Cap /proc/self/status (from inside) 1104846169 M * sannes got a nice oops on stopping a vserer now.. I suspect it is unmounting with quota bug .. 1104846202 M * Bertl could be ... 1104846210 M * Bertl (please upload if possible) 1104846327 M * Doener sannes: with 2.6 kernels it's bcapabilities 1104846364 M * Doener http://www-user.tu-chemnitz.de/~ensc/util-vserver/doc/conf/configuration.html 1104847344 J * infowolfe_ ~infowolfe@209-112-213-97-cdsl-rb1.nwc.acsalaska.net 1104847345 Q * infowolfe Read error: Connection reset by peer 1104848008 Q * infowolfe_ Quit: Leaving 1104848895 M * sannes ah, only have a snapshot of it on my cam from those panics. I'll upload them in a sec. 1104848980 J * infowolfe ~infowolfe@209-112-217-235-cdsl-rb1.nwc.acsalaska.net 1104849133 M * sannes Bertl : should be at http://www.sannes.org/panics/ now .. 1104849528 M * ndim kernel/built-in.o(.text+0x2425a): In function `vxh_dump_history': 1104849528 M * ndim : undefined reference to `smp_send_stop' 1104849536 M * ndim Hmm. 1104849559 M * ndim (that is 2.6.10-vs1.9.3.16) 1104850423 Q * sannes Read error: Connection reset by peer 1104850495 Q * weasel Quit: connecting to -ssl ircs.oftc.net 1104851097 J * sannes ~ace@home.skarby.no 1104851106 M * sannes hm... did i miss anything? 1104851754 M * sannes hm, got a panic on vserver start now.. heh entertaining.. 1104851767 M * Bertl hmm, could you upload it too? 1104851777 M * sannes I think this is the worst week ever for me, never had that many panics before.. yip, sure will.. 1104851848 M * sannes don 1104851850 M * sannes done 1104851852 M * sannes even :) 1104852035 Q * infowolfe Quit: Leaving 1104852652 J * Mega\work ~Megabart2@host111-101.pool80182.interbusiness.it 1104852655 M * Mega\work Hi 1104852677 M * Mega\work I have one problem when i create a new vserver 1104852714 M * Mega\work i've using this command for create a vserver 1104852723 M * Mega\work vserver name build 1104852741 M * sannes what utils and what vserver version ? 1104852805 M * sannes Doner : thanks for docs btw :) 1104852819 M * sannes Bertl : quota works fine atleast :) 1104852829 M * Doener np 1104852835 M * Mega\work vserver 1.29 1104852851 M * Mega\work the error is 1104852859 M * sannes Bertl : atleast with SYS_ADMIN 1104852884 M * Mega\work can't set the ipv4 root (function is not imoplemented) 1104852902 M * sannes Mega\work : have you booted on a vserver enabled kernel? 1104852905 M * Doener did you actually boot a vserver kernel? 1104852913 M * Doener sannes: you win ;) 1104852915 M * Mega\work yess 1104852946 M * Mega\work I have patched the kernel and i have recompiled the kernel 1104852949 M * Doener check with uname -a please ;) 1104852996 M * sannes Doener : lol :) 1104853079 A * Doener goes on food hunt, will be back later ;) 1104853082 N * Doener Doener|gone 1104853287 M * Mega\work How can reslv this problem? 1104853305 M * Bertl well, try with 'uname -a' for a start ;) 1104853356 M * Mega\work ok 1104853407 M * sannes (and paste the output here) 1104853539 M * sannes Mega\work : and paste the output here? 1104853599 M * Mega\work Linux debian 2.4.28-grsec #12 SMP Fri Dec 3 10:55:46 CET 2004 i686 unknown 1104853631 J * infowolfe ~infowolfe@209-112-217-235-cdsl-rb1.nwc.acsalaska.net 1104853732 M * Mega\work this is the output 1104853755 M * sannes and you did apply the patch? 1104853761 M * sannes (the vserver one) 1104853787 M * Mega\work one moment......... 1104853797 M * Bertl Mega\work: looks like you patched grsec but not vserver ... 1104853814 M * Mega\work I used this patch 1104853815 M * Mega\work grsecurity-2.0.2-2.4.28-vs1.29.patch 1104853834 M * Bertl url? 1104853858 M * Bertl nevermind, I found it ... 1104853902 Q * we2by Ping timeout: 480 seconds 1104853922 M * Bertl --- linux-2.4.28-vs1.29/Documentation/Configure.help 2004-11-23 18:05:45.128456128 -0600 1104853925 M * Bertl +++ linux-2.4.28-vs1.29-grsec/Documentation/Configure.help 2004-11-23 18:06:54.285942592 -0600 1104853926 J * we2by ~we2by@145.116.4.193 1104853936 M * Bertl this is a grsec patch _ontop_ of the vserver patch ... 1104853950 M * Bertl Mega\work: so you have to apply vs1.29 _first_ then the grsec 1104853975 M * Mega\work http://www.firehead.org/~jeffrey/linux-vserver/ 1104854070 M * Mega\work patch-2.4.28-vs1.29.diff 1104854083 M * Bertl yep, that one first ... 1104854094 M * Mega\work I must patch with patch-2.4.28-vs1.29.diff and after with grsecurity-2.0.2-2.4.28-vs1.29.patch? 1104854104 M * Bertl yes, exactly ... 1104854120 M * Mega\work ok 1104854234 M * sannes Bertl : did you get anything from the panics? 1104854253 M * Bertl roughly 30% until now ;) 1104854305 M * sannes Bertl : because I have added a couple of patches from bkbits that should have fixed a race condition on umount, but I wasn't able to reproduce it without vserver stop.. so was just wondering if I should try harder or if you found something.. :) 1104854337 M * sannes but then it happen on mount so..heh, worried me.. heh 1104854362 M * Bertl I'll look into it in a few minutes ... 1104854381 A * sannes is dreaming of a stable server with 2.6 vserver,vroot and bme running.. heh, . and thanks! :) 1104854394 M * sannes thanks for looking into it :) 1104854408 M * Bertl my pleasure! 1104854562 M * Bertl k, 355 looks strange but not vserver related, 356 did happen at boot time, right? 1104854697 M * Bertl sannes: is 357 and 358 with 1.9.3.16+fix? 1104855083 M * sannes 355 is old, probably because of ACPI thingy.. so ignore that 1104855095 M * sannes 356 happened on vserver stop 1104855109 M * sannes so did 357 .. and 358 on vserver start.. IIRC 1104855133 M * Bertl with 1.9.3.16+fix? 1104855135 A * sannes was like "oh shit, what are those numbers... heh". 1104855140 M * sannes yes, all with 1.9.3.16+fix 1104855228 M * Bertl well, I would suggest two things: a) use a serial console and b) avoid reiser and preemption ... 1104855245 M * sannes hehehe 1104855257 M * sannes can do on preemption atleast.. :) 1104855278 M * Bertl if the serial console is not an option, please select a screen font/resolution which allows for more chars 1104855287 M * sannes just did :) 1104855293 M * sannes but the font is really ugly 1104855295 M * Bertl (in both oopses is the important part missing ;) 1104855305 M * sannes hehe .. figures 1104855321 M * ndim 1.9.3.16 needs a fix? 1104855330 M * ndim I was justa bout to boot it. 1104855337 M * Bertl yep, a bug sneaked into it ... 1104855367 M * sannes :> but you'll be fine aslong as you don't use any netwo..... hehe 1104855381 M * Bertl nope preempt is the magic word ;) 1104855410 M * Bertl but give me a second and I'll upload a patch 1104855431 M * sannes hehe, yeah that'll work too... 1.9.3.17! :) 1104855480 M * Bertl you wish! 1104855505 M * Bertl http://vserver.13thfloor.at/Experimental/delta-2.6.10-vs1.9.3.16-vs1.9.3.16.1.diff 1104855525 Q * we2by Ping timeout: 480 seconds 1104855525 J * we2by ~we2by@dc5146d009.adsl.wanadoo.nl 1104855531 M * sannes hehe :) ok, going to make a better screenshot this time.. back in a sec.. hm, didn't have a serial cable here.. hm 1104855596 M * sannes this is just so much fun, now it didn't panic.. hey, I could film it any you play it of in slow-motion. :> 1104855635 M * ndim OK, and I have to get rid of preemption? 1104855652 M * Bertl well, it can't hurt to get rid of preemption anyway ... 1104855782 M * ndim OK, building again... 1104855786 M * sannes Bertl : don't you want to know about those bugs? I mean, yes, not in a production enviorment, but on my laptop .. 1104855831 M * Bertl sure I want to know, if they are linux-vserver related ... 1104855885 M * Bertl but a lot of them are not ... for example, Linus discovered that ACPI/ATI issue ... which caused ndims laptop to panic ... 1104855892 M * sannes and mine 1104855897 M * Bertl probably ... 1104855904 M * sannes got a ATI Mobility .. heh 1104855907 M * sannes radeon 1104855928 M * Bertl and I suspect a lot more issues with laptops ... 1104855929 M * ndim exactly. 1104855944 M * we2by I have about 4 vservers to boot at startup 1104855950 M * we2by I getthe following error 1104855951 M * Bertl doesn't mean that I do not care about the non-linux-vserver issues ... 1104855965 M * Bertl but they are second priority ... 1104855979 M * we2by ld SV respawning too fast: disabled for 5 min 1104856003 M * sannes I understand where you want to focus, but the mini bug in vserver wouldn't have taken a little bit more time to discover .. ? 1104856055 M * Bertl we2by: looks like you are starting init (probably in a vserver) and it tries to create/access/start something on a terminal/device/etc which isn't there -- it gives up after some time ... 1104856080 M * we2by Bertl, how do I fix it? 1104856110 M * Bertl sannes: yes, never said (and probably never will say) please don't report any bugs/panics to me ... 1104856147 M * Bertl the oposite is true, bring them to me, but do not _always_ expect me to (re-)solve them ;) 1104856179 M * Bertl we2by: probably by consolidating the inittab inside the vservers? 1104856211 M * we2by Bertl, I wish I can understand what u mean 1104856266 M * Bertl try 'grep respawn /etc/inittab' 1104856273 M * we2by in my host OS 1104856274 M * we2by ? 1104856295 M * Bertl well, you can try it there too ... you'll see a list of 'respawning' services 1104856307 M * we2by and? 1104856324 M * Bertl such services if they can not be started, produce your message 1104856441 M * we2by ok, I will take a llok at it tonight 1104856448 M * we2by have to code now 1104856613 M * sannes Bertl : by all means, I'm for one is very pleased with the service you provide (and not taking money for) :) and I can see that all code isn't up for preemption yet.. heh :) 1104856673 M * sannes and I'd hate to see you bogged down with bugs that isn't in your field of intrestæ 1104856700 M * sannes intrest/focus/expertise or whatever :) 1104856713 M * Bertl more focus/expertise ;) 1104856933 M * Bertl ndim: hmm, you are compiling up (to get that smp_* error, right?) 1104857015 Q * infowolfe Ping timeout: 480 seconds 1104857055 M * Bertl sannes: http://vserver.13thfloor.at/Experimental/diff-2.6.10-vs1.9.3.13-vr0.21.diff <-- should provide the CAP_QUOTACTL and use it (applies fine to 1.9.3.16.1) 1104857077 M * Bertl and thanks for testing that stuff ;) 1104857581 M * ndim Bertl: the smp error was while compiling 1.9.3.16 without SMP support. 1104857605 M * Bertl yep, thought so, will be fixed in the next release ... 1104857773 M * sannes Bertl : got two new panics for you, the last one is most intersting I think.. 1104857790 M * sannes nevermind! heh.. 1104857797 M * sannes picture is broken.. and blurred.. heh 1104857816 M * sannes but if you have a vivid imagination ... 1104857819 M * sannes :> 1104857825 M * sannes two secs 1104857870 J * infowolfe ~infowolfe@209-112-215-212-cdsl-rb1.nwc.acsalaska.net 1104857938 M * Bertl vivid imagination indeed ... 1104857973 A * Bertl is trying to 'read' The Matrix 1104858016 M * Bertl sannes: you should be able to activate framebuffer console and use a decent 8x8 font on 1024x768 or higher ... 1104858052 M * Bertl (that means roughly 120x96 chars) 1104858074 M * sannes take a look at it now.. 361 1104858168 M * Bertl well, I'd say that is a classical reiser quota bug ;) 1104858192 M * Bertl this is somewhat reproduceable with your current setup? 1104858248 M * matti Bertl: :D 1104858259 M * sannes Bertl : yes, everytime :) 1104858277 J * DuckMaster ~Duck@dyn-83-157-183-246.ppp.tiscali.fr 1104858281 M * Bertl every time you do what exactly? 1104858289 M * sannes Bertl : but, then again I think maybe those new patches in bk that did those quota changes could be blamed.. except I can't reproduce it on the commandline.. 1104858312 M * sannes stop the vserver that mounts .. 1104858325 M * sannes (and I tried without the usrquota,grpquota stuff aswell) 1104858327 M * Bertl k, you need at least the namespace stuff to make it happen 1104858362 M * Bertl it looks to me like the 'final' umount on the (quota enabled?) reiserfs kills the machine ... 1104858407 M * Bertl do you have a chance to try with a different fs (but everything else unchanged)? 1104858422 M * sannes I can do that, two min.. 1104858705 Q * DuckKing Ping timeout: 480 seconds 1104858828 M * sannes got a panic there aswell.. using ext2 1104858844 M * Bertl please upload ... 1104858955 M * sannes done 1104858976 M * sannes yes, yes .. I know.. this is where I turn of preempt .. 1104858979 M * Bertl and please try to enable spinlock debugging (2 options IIRC) in the kernel and rebuild ... 1104859037 M * Bertl argh! I'm blind ... :/ 1104859051 M * ndim ok, running 2.6.10-vs1.9.3.16.1 now. 1104859106 M * Bertl sannes: how comes that the oops in 362 is still in reiserfs_sync_fs() ? 1104859215 M * sannes whuut?! 1104859238 M * sannes arargh 1104859279 M * sannes I just want to do one little thing before I rebuild.. going to test 1.9.3.14 (because it worked) and see if has the same problems.. could be it is all my fault you know.. 1104859331 M * Bertl yes, 1.9.3.14 with preempt/reiser/etc would be interesting ... 1104859399 M * Bertl (it looks like reiser is somehow releasing the BKL without acquiring it first (not necessarily reiser related but chances are good) 1104859519 M * sannes http://www.sannes.org/panics/imgp0364.jpg <-- was the last one, btw 1104859535 M * sannes (not reiserfs_sync_fs()) heh.. copied the wrong file 1104859599 M * Mega\work How can use vserver with a dhcp? 1104859628 M * Bertl where is the dhcp and what is it supposed to do? 1104859770 M * Mega\work I must to use dhcp because in this lan i have a dhcp server 1104859946 M * Mega\work root@LDAP:/# ifconfig 172.16.0.110 1104859946 M * Mega\work 172.16.0.110: error fetching interface information: Device not found 1104859946 M * Mega\work root@LDAP:/# ifconfig eth0 172.16.0.110 1104859946 M * Mega\work SIOCSIFADDR: Permission denied 1104859946 M * Mega\work SIOCSIFFLAGS: Permission denied 1104859947 M * Mega\work root@LDAP:/# 1104859994 M * Mega\work Why i've this error? 1104860028 M * Bertl because you are not allowed to change the interface config from inside a vserver (on 2.4/1.2x and 2.6/1.9x w/o ngnet) 1104860094 M * Bertl also the dhcp assignment will not work as you expect it, because you have only one mac/nic and the dhcp server will not give you more than one address ... 1104860117 M * Mega\work How can resolv a problem? 1104860139 M * Mega\work I don't follow you 1104860142 M * sannes Bertl : can't crash it with 1.9.3.14 .. 1104860165 M * Mega\work the problem is in a confg file of vserver? 1104860179 M * Bertl sannes: k, I'll prepare a patch between .14 and .16 for testing 1104860180 M * Mega\work or is another problem? 1104860197 M * Bertl look, dhcp uses the mac address, right? 1104860201 M * Mega\work yess 1104860210 M * Bertl okay, how many mac addresses do you have? 1104860217 M * Mega\work ahhhhh 1104860243 M * Mega\work then.......in real server i have one interface 1104860257 M * Mega\work eth0 Link encap:Ethernet HWaddr 00:48:54:1C:9A:10 1104860257 M * Mega\work inet addr:172.16.0.104 Bcast:172.16.0.255 Mask:255.255.255.0 1104860257 M * Mega\work UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 1104860257 M * Mega\work RX packets:1031 errors:0 dropped:0 overruns:0 frame:0 1104860257 M * Mega\work TX packets:587 errors:0 dropped:0 overruns:0 carrier:0 1104860259 M * Mega\work collisions:11 txqueuelen:1000 1104860259 M * Mega\work RX bytes:96257 (94.0 KiB) TX bytes:75415 (73.6 KiB) 1104860261 M * Mega\work Interrupt:9 Base address:0x2000 1104860269 M * Mega\work this is the ifconfig of real server 1104860286 M * Bertl yep, and this one can (and probably does) work with dhcp 1104860286 M * TheSeer ip addr would have been less lines 1104860339 M * Bertl Mega\work: the vservers will use aliases (or at least addresses) on the same interface ... so no way to get them via dhcp 1104860342 M * Mega\work yes 1104860374 M * Mega\work I don't have an alias in a real server 1104860380 M * Bertl but you can use local/private addresses for your vservers, and SNAT them to the one 'official' address 1104860410 M * Mega\work this is the ifconfig of vserver LDAP 1104860411 M * Mega\work eth0 Link encap:Ethernet HWaddr 00:48:54:1C:9A:10 1104860411 M * Mega\work UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 1104860411 M * Mega\work RX packets:1169 errors:0 dropped:0 overruns:0 frame:0 1104860411 M * Mega\work TX packets:675 errors:0 dropped:0 overruns:0 carrier:0 1104860411 M * Mega\work collisions:11 txqueuelen:1000 1104860413 M * Mega\work RX bytes:107765 (105.2 KiB) TX bytes:86599 (84.5 KiB) 1104860414 M * Mega\work Interrupt:9 Base address:0x2000 1104860442 M * Mega\work well 1104860449 M * Bertl kernel and tools? 1104860460 M * Bertl (which version, I mean) 1104860479 M * Mega\work I have created the vserver first to patch kernel 1104860489 M * Mega\work when i delete a vserver create? 1104860535 M * Bertl well, when you created your vserver doesn't really matter ... 1104860683 M * Mega\work whem i created the vserver, i have had more error 1104860693 M * Mega\work the error was 1104860735 M * Mega\work can't set the ipv4 root (function is not imoplemented) 1104860779 M * Mega\work How can delete the vserver that I've created first? 1104860845 M * sannes Bertl : what can you prepare? should I turn on more debugging as you suggested? 1104860901 M * Bertl spinlock and spinlock_sleep debugging would not hurt ... 1104860915 M * Bertl Mega\work: rm -rf 1104860956 M * Mega\work rm: cannot unlink `ldap/proc/15586/exe': Permission denied 1104860956 M * Mega\work rm: cannot unlink `ldap/proc/15586/mounts': Permission denied 1104860956 M * Mega\work rm: cannot remove directory `ldap/proc/15586': Operation not permitted 1104860956 M * Mega\work rm: cannot remove directory `ldap/proc': Device or resource busy 1104860956 M * Mega\work rm: cannot remove directory `ldap/dev/pts': Device or resource busy 1104860957 M * Mega\work rm: cannot remove directory `ldap/dev': Directory not empty 1104860959 M * Mega\work rm: cannot remove directory `ldap': Directory not empty 1104860975 M * Bertl well, stopping it first would help ;) 1104860982 M * Mega\work ahahahahahahah 1104860995 M * Mega\work I'm more stupid :D 1104861049 M * Mega\work Bertl: for create vserver i use this command 1104861062 M * Mega\work vserver ldap build 1104861068 M * Mega\work It's right? 1104861099 M * Bertl well, that answers one of my questions about the tools you are using .. but not what exact version :/ 1104861132 M * Mega\work vserver is 1.2.9 version 1104861205 M * Bertl yep, but you seem to use util-vserver ... most likely version 0.30 or maybe 0.30.196? 1104861251 M * Mega\work i have installed the util server with a debian package 1104861307 M * Mega\work but I don't knowledge the version of util-vserver 1104861320 M * Bertl http://vserver.13thfloor.at/Stuff/SCRIPT/testme.sh 1104861328 M * Bertl it will tell you ;) 1104861453 Q * _are_ Quit: Disconnecting 1104861559 M * Mega\work inux-VServer Test [V0.07] (C) 2003-2004 H.Poetzl 1104861559 M * Mega\work chcontext is working. 1104861559 M * Mega\work chbind is working. 1104861559 M * Mega\work Linux 2.4.28-vs1.29-grsec i686/0.30/0.30 [J] 1104861559 M * Mega\work --- 1104861560 M * Mega\work [001]# succeeded. 1104861562 M * Mega\work [011]# succeeded. 1104861564 M * Mega\work [031]# succeeded. 1104861567 M * Mega\work [101]# succeeded. 1104861569 M * Mega\work [102]# succeeded. 1104861571 M * Mega\work [201]# succeeded. 1104861573 M * Mega\work [202]# succeeded. 1104861603 M * Bertl well, the tools are somewhat outdated, but they will work 1104861866 M * Mega\work Bertl: one question........you are a vserver development? 1104861915 M * Bertl guess I am ... 1104861929 M * ndim hehe 1104861954 M * Mega\work ??????? 1104861975 M * Bertl yes I am a linux-vserver developer! 1104861996 M * ndim Mega\work: Bertl is *THE* linux-vserver developer. 1104862043 M * Mega\work excuse me for the bad english :( 1104862074 M * Bertl no problem, I'm no native english speaker either ... 1104862075 Q * sannes Read error: Connection reset by peer 1104862098 M * Mega\work What do you speak? 1104862145 M * Bertl C, x86 asm, ppc asm, PHP, german ... 1104862163 M * daniel_hozac lol 1104862222 J * sannes ~ace@home.skarby.no 1104862239 M * sannes Bertl : enabled those two debug options.. and.. it paniced on bootup once, and rebooted when I tried to stop a vserver.. 1104862262 M * Bertl got a snapshot of the first panic? 1104862262 Q * sannes Read error: Connection reset by peer 1104862313 M * Mega\work panic 1104862320 M * Mega\work Bertl is the oracle :D 1104862336 J * sannes ~ace@home.skarby.no 1104862356 M * sannes Bertl : enabled those two debug options.. and.. it paniced on bootup once, and rebooted when I tried to stop a vserver.. 1104862371 M * Bertl hmm, deja vu: got a snapshot of the first panic? 1104862398 M * sannes it is all too long so you are missing the first part.. 1104862404 M * sannes don't know if it'll help you.. 1104862411 M * Bertl we'll see ... 1104862432 M * Bertl btw, here is a small patch: http://vserver.13thfloor.at/Experimental/delta-2.6.10-vs1.9.3.16.2_revnet01.diff 1104862456 M * Bertl you should apply it with patch -p1 -l -R 1104862465 M * Bertl (maybe --dry-run first) 1104862502 J * _sebd ~sebd@lesdeveloppementsdurables.org 1104862513 M * Bertl welcome _sebd! 1104862515 Q * sebd Read error: Connection reset by peer 1104862533 M * _sebd hi 1104862540 N * _sebd sebd 1104862557 M * sannes .. http://vserver.13thfloor.at/Experimental/delta-2.6.10-vs1.9.3.16.2_revnet01.diff .. for me? 1104862628 M * sannes :) 1104862658 M * Mega\work thanks all 1104862674 M * Bertl you're welcome! 1104862678 M * Mega\work i stop to work 1104862680 M * Mega\work :) 1104862694 M * Bertl so you become Mega\nowork? 1104862715 M * Mega\work I return it tomorrow 1104862729 M * Bertl k, cya! 1104862815 Q * sannes Read error: Connection reset by peer 1104863068 Q * Mega\work Quit: Leaving 1104863183 J * sannes ~ace@home.skarby.no 1104863189 M * sannes Bertl : still same problem.. 1104863200 M * Bertl ah, that sounds good then ;) 1104863215 M * sannes hehe :) I can't imagine why... 1104863236 M * sannes now we know atleast one place it doesn't go wrong.. heh 1104863276 M * Bertl hmm, not necessarily (at least from the logic point of view) 1104863317 M * sannes anyways, noticed a little oddity now.. 1104863335 M * Bertl hmm? 1104863359 M * sannes sometimes when I stop a vserver .. afterwards the volume is not mountable for anyone else, saying it is already mounted.. 1104863383 M * sannes I wonder if this has anything with me putting SYS_ADMIN in.. 1104863391 M * sannes *checking it out* 1104863402 M * Bertl well, I have a good idea what might be to blame ... 1104863442 M * Bertl but a) this will need some more investigation (on my part) and b) I have to leave in a few minutes 1104863506 Q * sannes Read error: Connection reset by peer 1104863624 M * Bertl k, back later ... 1104863632 N * Bertl Bertl_oO 1104864114 J * sannes ~ace@home.skarby.no 1104864129 M * sannes did I miss anything? last thing I saw was a) .. 1104864147 M * Doener|gone 19:33:44 Bertl k, back later ... 1104864147 M * Doener|gone 19:33:52 * Bertl is now known as Bertl_oO 1104864150 N * Doener|gone Doener 1104864463 M * sannes thanks 1104864832 Q * dsanta Ping timeout: 480 seconds 1104864857 M * ndim How does one go about creating the stuff for /etc/vservers/SERVERNAME/ ? 1104864907 J * _are_ ~are@84.56.137.187 1104865132 M * Doener ndim: if you're starting from scratch "vserver build -m skeleton ..." is a good choice, otherwise any editor should do ;) 1104865213 M * ndim Doener: Thanks! 1104865466 M * _are_ hi 1104865537 J * dsanta ~santa@c68.190.156.105.roc.mn.charter.com 1104865857 M * we2by santa? 1104865868 M * we2by I didn;t know santa also use vservers 1104866545 Q * albeiro Ping timeout: 480 seconds 1104866715 J * albeiro albeiro@linux.gentoo.pl 1104867122 Q * Kostko Remote host closed the connection 1104869968 M * we2by does some one have a slackware10 image? 1104870795 M * mugwump try the lycos images 1104871930 Q * sannes Read error: Connection reset by peer 1104872760 Q * serving Ping timeout: 480 seconds 1104873740 J * infowolfe_ ~infowolfe@209-112-215-212-cdsl-rb1.nwc.acsalaska.net 1104873740 Q * infowolfe Read error: Connection reset by peer 1104874634 N * Doener Doener_zZ 1104874637 M * Doener_zZ night! 1104874926 J * tanjix tanjix@pD9FAC35E.dip.t-dialin.net 1104875057 J * sannes ~ace@home.skarby.no 1104875480 Q * tanjix Quit: 1104876431 J * JonB ~NoSuchUse@kg99.kollegiegaarden.dk 1104877555 Q * infowolfe_ Ping timeout: 480 seconds 1104879484 J * serving ~serving@213.186.174.191 1104880158 N * Bertl_oO Bertl_zZ 1104880451 T * services.oftc.net http://linux-vserver.org/ | latest stable 1.29, devel 1.3.9, 1.9.3, ng8.7 1104880451 F * ChanServ -o Doener_zZ 1104880517 Q * JonB Quit: Leaving 1104880700 Q * serving Ping timeout: 480 seconds