1122340563 Q * monrad Remote host closed the connection 1122340625 J * monrad ~monrad@213083190130.sonofon.dk 1122341810 Q * tchan Ping timeout: 480 seconds 1122343490 Q * anonymousc Ping timeout: 480 seconds 1122345444 N * Bertl_oO Bertl 1122345462 M * Bertl morning folks! 1122346250 M * micah goodmorning Bertl! 1122346292 M * Bertl hey micah! LTNS! 1122346313 M * micah yeah, I've been in finland, and then recovering! 1122346382 M * micah how goes 2.0? 1122346509 J * ddlp_ ~daniel@205.234.152.200 1122346518 M * Bertl well, we are at rc8.1 ... so far so good :) 1122346519 Q * ddlp Read error: Connection reset by peer 1122346554 M * Bertl FaUl: still not around? 1122346602 M * micah Bertl: wow, great! 1122346621 M * Bertl and enrico released a new version too! 1122346645 M * micah Bertl: I was just beginning to start trying to port the FOR patches to debian's recently uploaded 2.6.12 1122346687 M * Bertl hmm, doesn't the rc8.1 apply anyway? 1122346778 M * micah Bertl: I was going to try that first 1122346778 M * Bertl btw, we now have an official report that it works on xen too :) 1122346789 M * micah oh, thats... weird! 1122346822 M * Bertl well, no, actually it's just another arch we support :) 1122346853 M * Bertl and guess what, we now have all syscall numbers _including_ the parisc one, yay! 1122346867 M * micah wow!! 1122346939 M * micah I guess I should try xen one of these days 1122347134 M * Bertl yeah, just don't expect linux-vserver grade stability, arch coverage or quality of support, otherwise you will be disappointed :) 1122347245 M * micah rc8.1 applies without any rejects, only a little off on: 1122347245 M * micah patching file arch/ppc/Kconfig 1122347246 M * micah Hunk #1 succeeded at 1350 (offset 18 lines). 1122347249 M * micah but I can fix that 1122347260 M * Bertl ahe, that was already fixed :) 1122347570 M * Aiken when compiling the tools against glibc is it possible to compile them static? 1122347597 M * Bertl hmm .. maybe, but will for sure give huge binaries 1122347626 M * Aiken does not matter 1122347628 M * Bertl Aiken: do you still see any issues with dietlibc? 1122347650 M * Aiken have not had anymore problems 1122347738 M * Aiken the cross tool chain I just built has a different glibc version to what is on the sparc so want static binaries 1122347762 M * Bertl hmm, dietlibc is very cross compile friendly 1122347784 M * Bertl i.e. you 'just' compile it for that specific arch, the rest will be done automagically 1122348053 M * Aiken ok, having a play 1122348562 N * ddlp_ ddlp 1122348980 M * Aiken seemed to work 1122349122 M * Bertl excellent 1122349147 M * Bertl default 0.30.208 build I presume? 1122349158 M * Aiken yes 1122349182 M * Aiken even if that kernel is dodgy I thought I would at least try testme.sh 1122349215 M * Aiken maybe a bit too dodgy, just rebooting after a long fsck :( 1122349986 M * Bertl any files left? 1122350211 M * Aiken before the case I had installed a binary package of the tools 1122350219 M * Aiken after the crash they were there in name only 1122350241 M * Aiken directory entries an no contents 1122350257 M * Bertl which fs? 1122350285 M * Aiken ext2 1122350320 M * Aiken when I am finished playing I will be formatting and restoring from backup 1122350355 M * Bertl makes sense! thanks for your testing, btw!! 1122350551 M * Bertl it's fascinating how much easier a panic is to read (on hppa) if you enable the console :) 1122350590 M * Aiken I wish a normal pc had a proper console 1122350618 M * Bertl yeah, well, we are ages from there, although iLo is quite close 1122350647 M * Aiken a friend in IT at the local uni loves remote consoles 1122350655 M * Aiken my experience is the serial console on sparc 1122350669 M * Aiken he raves about network based consoles on some of the machines he manages 1122350716 M * Bertl yeah, well, the parisc console is quite powerful .. although the command interface is ancient 1122350786 M * Aiken http://pastebin.com/320773 1122350821 M * Aiken the machine is not happy 1122350826 M * Bertl hmm, do you have bash too? 1122350866 M * Aiken trying that now, for scripts with out +x I just use sh 1122350885 M * Bertl ah, and btw, you have an old testme.sh 1122350886 M * Aiken much better http://pastebin.com/320774 1122350910 M * Bertl yeah, looks much better :) 1122350941 M * Bertl now we are at the point where I asked you to test something for me *G* :) 1122350956 J * tchan ~tchan@c-24-13-81-164.hsd1.il.comcast.net 1122350968 M * Bertl but it's probably trivial compared to the hoops you jumped through to get there ... 1122350977 M * Bertl wb tchan! 1122351178 M * Bertl could you download the http://vserver.13thfloor.at/Experimental/SYSCALL/syscall_shiny5.h and replace the file lib/syscall-alternative.h of the tools, then run configure again (it should pick up the alternative syscall method) and test if the tools work? 1122351423 M * Aiken grabbed the file, wife is waiting for me because I said I would go shopping 1122351433 M * Aiken I'll try to have a look in a few hours 1122351442 M * Bertl well, don't let her wait! set priorities! 1122351540 M * Aiken how many arch to test it on? 1122351578 M * Bertl well, it was already tested (successfully) on alpha, x86, x86_64 .. and is soon to be tested on parisc and s390 1122351604 M * Aiken a real s390 or hercules? 1122351606 M * Bertl no sparc, mips, arm or m68k yet ... 1122351613 M * Bertl hercules of course ... 1122351620 M * Aiken oh well 1122351644 M * Aiken she is outside so I am out of here 1122351653 M * Bertl have fun! 1122351948 M * micah Bertl: what is a single \ for in the code? for example: 1122351960 M * micah #define VXH_HERE() \ 1122351960 M * micah ({ __label__ here; \ 1122351960 M * micah here:; \ 1122351960 M * micah &&here; }) 1122351969 M * micah continuation of a line? 1122351987 M * Bertl yep, it's a macro, and this one is multiline 1122352017 M * micah I maen, is it just for aesthetic reasons, using the \ 1122352026 M * micah or its something about C I dont know :) 1122352040 M * Bertl well, you could also write it into a single line ... 1122352061 M * micah ok, but if it were written on multiple lines without the \ it would be in error? 1122352069 M * Bertl yep, definitely 1122352095 M * Bertl it would #define VXH_HERE() to be empty 1122352109 M * Bertl and add an unhandled expression with the rest 1122352128 M * micah ok 1122352593 M * micah sanity check... I'm taking debian kernel source 2.6.8 applying the latest vs195 patch against it, and now trying to apply each individual FOR-2.0 patches to that tree 1122352649 M * micah everything in /include/linux/vserver should be what patch-2.6.12.3-vs2.0-rc8.1.diff has, right? 1122352667 M * Bertl I'd say so ... yes 1122352703 M * Bertl same goes for /include/linux/vs_* and kernel/vserver/* I'd days 1122352708 M * micah I ask because if I apply the 1.9.5 patches, and then try to apply delta-debug-feat02.diff it fails because of some significant differences in debug.h between 1.9.5 and 2.0 1122352709 M * Bertl s/days/say/ 1122352735 M * micah so i am just not going to fix those, but instead take those changes from 2.6.12.3-vs2.0-rc8.1.diff 1122352753 M * micah yeah, anything that is added by the patch 1122352753 M * Bertl yeah, could work ... 1122352784 M * Bertl otoh, there should be no rejects to those files, no? 1122352833 M * micah there are: 1122352850 M * Bertl maybe you are missing some patches from: http://vserver.13thfloor.at/Experimental/FOR-1.9.6/ too 1122352879 M * micah ooooh 1122352881 M * micah I am 1122352889 M * micah I did not know about that piece 1122352901 A * micah goes back to the drawing board 1122352907 M * Bertl but be careful, some of them are not part of vs2.0-rc* 1122352953 M * micah hmm, how can I tell? 1122352995 M * Bertl banner is, blkio,bme,fault is not 1122353040 M * Bertl infohide is, init_vx I guess so, inodes yep 1122353071 M * Bertl kconfig don't remember, but I think yes 1122353112 M * Bertl namei yes, mounts yes, syslog yes 1122353143 M * Bertl the unhold fix in slightly modified version 1122353176 M * Bertl vkill yes, vroot too, vshelper clean yep 1122353195 M * Bertl x25 fix is a kernel fix, it was somehow included and requires a different 'fix' now 1122353250 M * micah ok, i'll start trying to apply these 1122353275 M * Bertl if you fail, then I probably already included the specific patch in the debian version 1122353292 M * Bertl (i.e. if patch suggests to use -R ) 1122353367 M * micah but only if it fails with the suggestions of -R, other failures need to be resolved, right? 1122353381 M * Bertl probably ... 1122355379 M * Hollow morning 1122355457 M * Bertl morning Hollow! 1122355494 M * Hollow Bertl: after creating the context, do i have to set the flags outside the context? 1122355565 M * Hollow because i always get EPERM if i do set_caps after ctx_create 1122355722 M * micah got FOR-1.9.6/delta-banner-feat01.diff and delta-infohide-feat01.diff cleaned up, now time for sleeping 1122355830 M * Bertl Hollow: how do you create it? 1122355855 M * Bertl btw, s390x just passed the tests with shiny5 :) 1122355871 M * Bertl micah: good night then! 1122356160 M * Bertl okay, guess I'm off to bed too ... 1122356178 M * Hollow Bertl: vx_create(123); 1122356184 M * Bertl Hollow: if you end the SETUP (flag) then you will not be allowed to change stuff from inside ... 1122356210 M * Hollow i did not set anything, just create a context 1122356217 M * Hollow and after that, set e.g. RAW:ICMP 1122356254 M * Hollow i can set caps from outside, but not from inside 1122356259 M * Bertl please enable the syscall switch debug, and show me the output 1122356307 M * Hollow well, have to recompile the kernel first... i have no debugging support atm 1122356324 M * Bertl okay, then I will look at it later ... 1122356330 M * Hollow yup, n8! 1122356340 M * Bertl if you enable debug output try with enricos tools and have a look at that 1122356345 M * Hollow ok 1122356347 M * Bertl compare it with your output ... 1122356377 M * Bertl okay, good luck then ... and have a nice whatever ... 1122356382 N * Bertl Bertl_zZ 1122356384 M * Hollow cya! 1122358342 Q * Hollow Remote host closed the connection 1122358355 J * Hollow ~Hollow@home.xnull.de 1122359789 N * bipsen_Zz bipsen 1122361250 J * erwan_ho ~erwan@konilope.dyndns.org 1122362696 M * meebey good morning 1122362703 M * meebey I need help about 2 things 1122362710 M * meebey Jul 26 06:25:01 backup_lincoln pam_limits[3886]: setrlimit limit #6 to soft=-1, hard=-1 failed: Operation not permitted; uid=0 euid=0 1122362717 M * meebey thats one, I am getting that since sarge 1122362745 M * meebey and the other one is the procaccess patch porting to thenew util-vserver 1122363512 Q * erwan_ho Remote host closed the connection 1122363540 J * erwan_ho ~erwan@konilope.dyndns.org 1122363566 Q * erwan_ho Remote host closed the connection 1122364414 J * prae ~prae@gut75-1-81-57-27-189.fbx.proxad.net 1122364545 J * Loki|muh loki@satanix.de 1122364643 Q * Loki|muh_ Read error: Connection reset by peer 1122366258 Q * monrad Ping timeout: 480 seconds 1122367911 T * * http://linux-vserver.org/ | latest stable 1.2.10, 2.0-rc8.1, ngnet ng9.5 -- 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 ;) 1122367911 T * Bertl - 1122367931 M * bipsen is it just me - or is there a problem accessing the linux-vserver.rog website right now ? 1122368052 M * Aiken is not just you, I can not get it either 1122368090 M * bipsen maybe the vserer ruuning the webserver has crashed :-P 1122368094 M * bipsen ;- 1122368095 M * bipsen ;-) 1122368234 M * bipsen okay, seems like it came back to life again... 1122368318 M * Aiken one of lifes many little mysteries 1122368623 J * monrad ~monrad@213083190130.sonofon.dk 1122368756 N * BobR_oO BobR 1122368970 M * meebey urgs 1122368972 M * meebey ECHAN 1122370366 M * meebey vserver-stat is buggy 1122370373 M * meebey CTX PROC VSZ RSS userTIME sysTIME UPTIME NAME 1122370373 M * meebey 0 59 2.9G 45.8K 2m34s72 1m41s63 19h09m18 root server 1122370378 M * meebey 2.9G is a bit much 1122370699 J * lilo_ ~lilo@lilo.usercloak.oftc.net 1122370699 Q * lilo Read error: Connection reset by peer 1122371809 M * meebey does someone know vserver internals? 1122371822 M * meebey I tried to port a patch that worked for the old util-vserver 1122371829 M * meebey but I was not successful 1122372024 N * BobR BobR_afk 1122373315 M * meebey hm show stopper is now the procaccess thing 1122373320 A * meebey retries 1122374567 Q * Aiken Read error: Connection reset by peer 1122376628 Q * albeiro Ping timeout: 480 seconds 1122376898 J * Aiken ~james@tooax6-231.dialup.optusnet.com.au 1122377130 J * albeiro ~albeiro@graffias.estrefa.pl 1122377451 J * Blogmeister ~Blogmeist@site.lycos.de 1122379307 N * BobR_afk BobR 1122379384 P * Blogmeister 1122381171 Q * Aiken Quit: Leaving 1122382361 N * BobR BobR_afk 1122382490 M * meebey Available APIs: compat,v11,v13,fscompat,net,oldproc,olduts 1122382499 M * meebey how can I found out what API I am using? 1122382505 M * meebey my vserver version is 1.2.10 1122382768 J * Loki|muh_ loki@satanix.de 1122382782 M * FaUl bipsen: now again, i'm right in the netherlands 1122382848 Q * Loki|muh Ping timeout: 480 seconds 1122382983 M * meebey galilei:/vservers# vserver vpn enter 1122382987 M * meebey Unknown flag 'procaccess' 1122382992 M * meebey hmpf 1122383332 Q * Loki|muh_ Read error: Connection reset by peer 1122383355 Q * nokoya Ping timeout: 480 seconds 1122383392 J * Loki|muh loki@satanix.de 1122383403 N * BobR_afk BobR 1122383463 M * meebey hmmmpf 1122383614 M * meebey http://paste.debian.net/1336 1122383622 M * meebey thats the patch I made, any one seeing the problem? 1122383631 M * meebey it's not recognizing the flag 1122383791 Q * cryo Quit: Покидаю 1122383831 J * cryo ~say@212.86.243.154 1122384564 M * meebey comparing: lock == lockcomparing: procaccess == lockcomparing: procaccess == schedcomparing: procaccess == nproccomparing: procaccess == privatecomparing: procaccess == fakeinitcomparing: procaccess == hideinfocomparing: procaccess == ulimitcomparing: procaccess == namespacecomparing: procaccess == procaccessUnknown flag 'procaccess' 1122384595 M * meebey so this didnt match?!? comparing: procaccess == procaccess 1122384611 M * meebey I have the feeling the check loop is buggy 1122384620 M * meebey but I didnt touch it 1122384901 J * nokoya ~young@hi-230-82.tm.net.org.my 1122385265 N * BobR BobR_afk 1122385538 M * meebey found the problem 1122385546 M * meebey some evil C hacker made this: 1122385547 M * meebey uint_least32_t 1122385547 M * meebey vc_text2cflag_compat(char const *str, size_t len) 1122385560 M * meebey max return value is 255 and 256 is 0 1122385562 M * meebey 0 == error 1122385563 M * meebey great 1122385573 M * meebey with the old util-vserver I could use 256 1122385796 M * meebey errr infact the char type in the flag list is the problem 1122385801 M * meebey making my 256 a 0 1122387464 J * Pazzo ~Pazzo@host130-250.pool8172.interbusiness.it 1122389514 M * meebey weee, my patch works now 1122389566 M * meebey after fixing unsigned char to uint_least32_t 1122389776 Q * mountie Remote host closed the connection 1122389939 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com 1122391989 M * pusling if two vservers wants to use as much cpu as possible, how is the cpu split between them ? 1122392193 J * erwan_ho ~erwan@konilope.dyndns.org 1122392215 M * meebey like between 2 normal processes 1122392220 P * erwan_ho 1122392231 M * meebey nice and priority have influence 1122392270 J * Doener ~doener@p5487678F.dip.t-dialin.net 1122392371 M * pusling okay 1122392436 N * Bertl_zZ Bertl 1122392444 M * Doener morning Bertl 1122392456 M * Bertl morning Doener! 1122392463 M * Bertl morning folks! 1122392569 M * Bertl FaUl: ping! 1122392717 M * meebey pffft 1122392723 M * meebey now everyone is back 1122392732 M * meebey after fighting with vserver internals 1122392774 M * Bertl meebey: well, what are you doing? 1122392792 M * meebey Bertl: I was porting the procaccess patch to the new util-vserver 1122392799 M * meebey Bertl: and I found the problem I had 1122392815 M * Bertl hmm? 1122392839 M * Bertl we are talking about util-vserver 0.30.208? 1122392848 M * meebey yes 1122392853 M * meebey I ported it from 0.29 1122392876 M * meebey Bertl: http://paste.debian.net/1340 1122392878 M * Bertl what do you need to port here? 1122392883 M * meebey Bertl: thats the ported patch 1122392903 M * meebey 0.29 was way different 1122392942 M * meebey since my patch didnt went in official I have to keep porting it 1122392960 M * meebey when bigger things are changed 1122393025 M * Bertl please remind me what the patch did? 1122393037 M * meebey its 0.3.204 btw but that should be a different for 0.29 -> 0.3.0. 1122393048 M * meebey Bertl: allows write access to the proc system 1122393057 M * meebey Bertl: required to use openswan inside a vserver 1122393116 M * meebey the old patch for 0.29 was: 1122393210 M * Bertl ah, okay, I remember now ... 1122393392 M * meebey http://paste.debian.net/1343 1122393404 M * meebey porting was easy except that ugly unsigned char! 1122393411 M * meebey who did that? I could kiss him :) 1122393422 M * meebey the function that returns the id of the flag is uint32 1122393431 M * meebey the listitself is unsigned char... 1122393440 M * meebey that made my next number 256 to a 0 1122393449 M * meebey 0 means error in that context for that function 1122393450 M * meebey cool eh? 1122393468 M * meebey so vserver vpn enter said, Unknown flag... 1122393493 M * meebey so the fix was to make the list uint32 too 1122393534 M * Bertl i.c. 1122393561 M * Bertl and the compiler didn't warn? 1122393632 M * meebey nope 1122393661 M * meebey C crap I call that 1122393679 M * meebey time to go, see ya 1122393789 M * Bertl k, cya 1122394150 M * Bertl WhatTheHack Sun. 31th of July 13:00 Linux-VServer 1122394183 M * Bertl FaUl: please get back to me, I could use some help ... 1122397360 M * Bertl okay, off for now .. back later this evening ... 1122397364 N * Bertl Bertl_oO 1122400319 J * revenger_ ~joe@bulldog.infosys.de 1122400713 Q * revenger Ping timeout: 480 seconds 1122400830 M * FaUl Bertl_oO: maybe this evening, is a lot of stress here 1122401794 Q * prae Quit: Execute Order 69 ! 1122402430 M * micah Bertl_oO: you are presenting at WTH? 1122402472 M * bipsen Anyone tried to check out the latest util-vserver from CVS ??? I have a problem trying to install a guest OS on a LVM partition: 1122402495 M * bipsen /usr/lib/util-vserver/vserver-build.functions: line 167: test: too many arguments 1122402533 M * bipsen line looks like: test isDirectoryEmpty "$VDIR" -a ! -e "$SETUP_CONFDIR" || { 1122403298 M * daniel_hozac isDirectoryEmpty "$VDIR" && test ! -e "$SETUP_CONFDIR" || { 1122403301 M * daniel_hozac i think. 1122404235 M * bipsen New error: /usr/lib/util-vserver/vserver-build.functions: line 167: test: isDirectoryEmpty: unary operator expected 1122404311 M * bipsen isDirectoryEmpty is a function defined in the functions file 1122404364 M * daniel_hozac what does line 167 look like now? 1122404441 M * bipsen test isDirectoryEmpty "$VDIR" && ! -e "$SETUP_CONFDIR" || { 1122404472 M * daniel_hozac the test should be after the && 1122404590 M * bipsen something better is happening ... (the expected is happening) ... now I hust need to test, that it also works, when the dest-dir ins't empty :-) 1122406248 M * bipsen Oups..... errors: 1122406248 M * bipsen vcontext: vc_create_context(): File exists 1122406249 M * bipsen vcontext: vc_create_context(): File exists 1122406249 M * bipsen vcontext: vc_create_context(): File exists 1122406255 M * bipsen a whole bunch of them 1122407574 M * bipsen I wonder if there's one for each rpm package installed on the host.... 1122409540 J * DaCa-fi ~danny@mail.limehouse.org 1122409735 J * erwan_ho ~erwan@konilope.dyndns.org 1122411834 Q * erwan_ho Remote host closed the connection 1122414732 N * bipsen bipsen_zZ 1122415713 J * Aiken ~james@tooax8-063.dialup.optusnet.com.au 1122418987 J * revenger ~joe@bulldog.infosys.de 1122419383 Q * revenger_ Ping timeout: 480 seconds