1122163426 M * bipsen Bertl: Hmm... well, I guess I'll have to live with the risk until the final cap-stuff is in place... no-one will have shell-access to that vserver anyway (except me) - so id anything should happen, it should be caused by a bug in Bind 9.2
1122163443 M * hwarrier I am getting this error when I stop a vserser ( http://pastebin.com/319391 ) - I kind of remember seeing something like this before and I just cant remember what I did before to fix it.
1122163531 M * Bertl hwarrier: could you give testme.sh a spin please? http://vserver.13thfloor.at/Stuff/SCRIPT/testme.sh 
1122163541 M * hwarrier debian with utils 0.30.207-8 on linux 2.6.12.3
1122163543 M * hwarrier k I will
1122163545 M * Bertl (and upload the output in the same place)
1122163566 A * Jani waves out. "Good night all."
1122163572 M * Bertl night Jani!
1122163582 Q * Jani Quit: 
1122163713 M * hwarrier bertl: it didnt complain anything ( http://pastebin.com/319394 )
1122163750 M * Bertl okay, let's update to rc8.1 and 0.30.208 and your issues will go away ...
1122164007 M * hwarrier hehe.. hadn't seen 8.1.. it comes higher in sort order :)
1122164030 M * Bertl yeah, it's funny ... you're not the first one falling for that
1122164036 M * hwarrier is 0.30.208 available in unstable/testing yet?
1122164075 M * Bertl no idea, you can get it here: http://www.13thfloor.at/~ensc/util-vserver/files/alpha/
1122164092 M * hwarrier thanks
1122164378 M * bipsen anyone online running Vserver on CentOS 4 or WBEL4 ?
1122164577 Q * hwarrier Ping timeout: 480 seconds
1122164591 M * bipsen During install I get: /bin/chown: cannot access `/var/lib/rpm/[A-Z]*': No such file or directory
1122164617 M * bipsen wonder where that one came from.... doing an "rpm -qa" inside the vserver afterwards doesn't return anything
1122164820 M * bipsen probably an issue regarding installation-order of the RPM's ... ?
1122164902 M * Bertl could be, or my favorite 'base-system' theory
1122164920 N * bipsen bipsen_Zz
1122164957 M * Bertl night then!
1122165001 M * bipsen_Zz night ... better catch some sleep - gotta be up to watch F1 on TV tomorrow (today ;-))
1122166403 J * hwarrier hwarrier@adsl-67-122-215-44.dsl.pltn13.pacbell.net
1122166448 M * hwarrier bertl: rc8.1 did it - it is working fine (even without 0.30.208). can I skip the update to 208? I like to wait for my apt-get to get it automatically later.
1122166635 M * Bertl well, if it works for you, keep it :)
1122166790 M * Bertl why change a running system when it works ... but if you encounter issues (e.g. with shutdown or reboot) please upgrade first ...
1122166933 Q * hwarrier Ping timeout: 480 seconds
1122172863 J * Doener` ~doener@p54876150.dip.t-dialin.net
1122173309 Q * Doener Ping timeout: 480 seconds
1122179326 Q * lilo Remote host closed the connection
1122180921 Q * pusling Read error: Connection reset by peer
1122180923 J * pusling_ ~pusling@195.215.29.124
1122182539 M * MooingLemur cat /proc/mdstat
1122182542 M * MooingLemur crap
1122182582 M * Bertl hmm ...
1122187345 J * lilo ~lilo@lilo.usercloak.oftc.net
1122191784 M * Hollow morning
1122192438 M * Bertl morning ... I'm off to bed now ...
1122192481 N * Bertl Bertl_zZ
1122193280 Q * ddlp Remote host closed the connection
1122194445 N * pusling_ pusling
1122195239 M * nokoya hello
1122195296 A * nokoya wanna ask something out of topic
1122195304 M * nokoya is 192.168.1.0/26 = 192.168.1.3-64 ?
1122195311 M * nokoya main ip is 192.168.1.2
1122195711 M * daniel_hozac 192.168.1.63 would be the broadcast address.
1122197772 N * bipsen_Zz bipsen
1122198764 M * bipsen question reharding the distributions folder.... rpmlist.d - this one is used, if the installation-method is RPM ?? how about the yum and apt folders ?
1122198873 M * bipsen and what does the --reinstall option in the 02 file in e.g. the fc03 folder mean ?
1122202236 N * cereal pg`aw|cereal
1122202642 J * war- ~war@lucidpixels.com
1122204420 J * prae ~benjamin@sherpadown.net
1122205404 Q * prae Quit: Pwet
1122211802 J * sannes ~ace@cm-84.118.218.175.chello.no
1122211830 J * gin_lit ~gin@gw-swsoft.ll-nsk.zsttk.ru
1122213712 Q * Doener` Quit: Leaving
1122213768 J * Doener ~doener@p54876150.dip.t-dialin.net
1122214928 M * Hollow hey Doener 
1122214932 M * Hollow long time no see
1122214934 M * Hollow ;)
1122215227 M * Doener yeah, been busy with life
1122216224 Q * daniel_hozac Quit: thunderstorms :|
1122216248 N * Bertl_zZ Bertl
1122216272 M * Bertl morning folks!
1122216297 M * Doener morning Bertl 
1122216320 Q * maharaja Remote host closed the connection
1122216330 M * Bertl Doener: busy with what?! :)
1122216367 M * Bertl nokoya: 192.168.1.0/26 = 192.168.1.0 ... 192.168.1.63
1122216387 M * Doener some strange thing called "life"... my girlfriend introduced me to it... it's strange, folks don't write what they want to tell you but make noises... really strange...
1122216423 M * Bertl Doener: any proof that what you claim is true?
1122216458 M * Bertl Hollow: btw, I'm going to rewrite the syscall.h once again :)
1122216477 M * Hollow Bertl: good, shiny3 fails with gcc4
1122216491 M * Doener don't think so... I guess I'm going insane and it's all just in my mind ;)
1122216493 M * Bertl ah, details?
1122216517 M * Hollow ../../lib/syscall.c: In function 'vserver':
1122216517 M * Hollow ../../lib/syscall.c:30: error: PIC register 'ebx' clobbered in 'asm'
1122216517 M * Hollow ../../lib/syscall.c:30: error: PIC register 'ebx' clobbered in 'asm'
1122216530 J * maharaja maharaja@80.64.143.52
1122216547 M * Hollow shiny2 works
1122216569 M * Bertl ah, look gcc _is_ getting smarter after all?
1122216574 Q * Doener Quit: Leaving
1122216579 M * Hollow :P
1122216599 M * Bertl welcome maharaja!
1122216755 M * bipsen Hi Bertl.... Yesterday you referred to your base-system theory... care to explain ? Currently my 01 list is: 
1122216759 M * bipsen basesystem
1122216759 M * bipsen glibc
1122216759 M * bipsen filesystem
1122216759 M * bipsen glibc-common
1122216759 M * bipsen libgcc
1122216761 M * bipsen setup
1122216761 M * bipsen tzdata
1122216775 Q * maharaja Remote host closed the connection
1122216807 M * bipsen anyway - through the major list of packages, I still end up with:/bin/chown: cannot access `/var/lib/rpm/[A-Z]*': No such file or directory
1122216830 M * bipsen which I seem to have located to the rpm package... could it be a missing dependency or something like that ?
1122216850 J * maharaja maharaja@80.64.143.52
1122216859 J * Doener ~doener@p54876150.dip.t-dialin.net
1122217060 M * bipsen to me it seems like there could be an issue with the link from /var/lib/rpm <-> /.rpmdb  ... who populates /.rpmdb - and when ?
1122217135 Q * Doener Read error: Connection reset by peer
1122217249 J * Doener ~doener@p54876150.dip.t-dialin.net
1122217419 Q * meebey arion.oftc.net quasar.oftc.net
1122217419 Q * gregster arion.oftc.net quasar.oftc.net
1122217419 Q * Medivh arion.oftc.net quasar.oftc.net
1122217419 Q * nox arion.oftc.net quasar.oftc.net
1122217484 J * nox ~nox@nox.user.oftc.net
1122217484 J * Medivh ck@paradise.by.the.dashboardlight.de
1122217484 J * gregster ~gregor@greart.de
1122217484 J * meebey meebey@booster.qnetp.net
1122218132 M * maharaja hi!
1122218145 M * maharaja does any1 of you use bind9 inside the vserver?
1122218157 M * maharaja because i compiled the server with-threads, without capabilities
1122218165 M * maharaja but it does not start and produces some weired error message
1122218471 M * Bertl let's hear ...
1122219042 J * flock ~restless@l192-117-111-12.broadband.actcom.net.il
1122219868 Q * Doener Quit: Leaving
1122219898 J * Doener ~doener@p54876150.dip.t-dialin.net
1122220228 M * maharaja Bertl: bind states that he is not able to do "named -u bind"
1122220240 M * maharaja and i need a kernel > 2.3.x or 2.2.22
1122220248 M * maharaja so 2.6 should suffice ;)
1122220704 M * maharaja mhm - i simly uncommented that damn error message
1122220707 M * maharaja so lets se what happens ;)
1122220931 M * Bertl hmm?
1122220951 M * Bertl maharaja: you lost me regarding the error and your 'solution?'
1122221297 M * sannes Bertl : I get vroot[0]_get_real_bdev: dev=dfedf380[dfedf3ec,253,29] everytime I use a quota tool (and I do that quite alot on this server) .. maybe drop that message?
1122221327 M * Bertl hmm, yeah .. sounds reasonable ...
1122221530 M * sannes could add a debug vroot option, but it has been rock stable for me..
1122221551 M * sannes (atleast in 2.4, only used it briefly on 2.6)
1122221572 M * Bertl well, 2.4 and 2.6 vroot is very different :)
1122221581 M * Bertl but yeah, I'll make that conditional
1122221592 M * Bertl (currently it is printed unconditional)
1122221642 M * sannes exactly :)
1122222395 Q * _ag_ Ping timeout: 480 seconds
1122222683 J * _ag_ ag@caladan.roxor.cx
1122222738 Q * _ag_ Quit: 
1122222745 J * _ag_ ag@caladan.roxor.cx
1122222759 M * maharaja Bertl: i removed the error and die message from the bind code ;)
1122222763 M * maharaja and now it starts without a problem
1122222775 Q * _ag_ Quit: 
1122222788 M * maharaja bertl: but it has got a problem with creating a listening interface
1122222792 M * Bertl aha, what exactly does it do/need?
1122222797 J * _ag_ ag@caladan.roxor.cx
1122222829 M * Bertl maharaja: are you sure that you are not working around a config/setup issue?
1122222845 M * maharaja honestly i do not know
1122222846 M * Bertl maharaja: maybe you have bind running (and binding) on the host already?
1122222909 M * maharaja http://raoul.bhatia.at/~raoul/bind9_log
1122222928 M * Bertl testme.sh output too?
1122222999 M * Bertl what does 'cat /proc/self/ninfo' show inside the guest?
1122223081 M * maharaja testme is all workgin
1122223091 M * Bertl fine, output?
1122223147 M * maharaja 1sec
1122223151 M * maharaja had to switch pc
1122223185 M * maharaja NID:    49161
1122223185 M * maharaja V4Root[0]:      80.64.143.42/255.255.255.224
1122223185 M * maharaja V4Root[bcast]:  255.255.255.255
1122223230 M * Bertl and the output of testme.sh please (via pastebin or so)
1122223322 M * maharaja http://pastebin.com/319699
1122223325 M * Bertl tx
1122223345 M * maharaja np
1122223356 M * Bertl please try with rc8.1 + 0.30.208
1122223383 M * maharaja so you think its vserver related?
1122223425 M * Bertl it could be, but if I dig into the code, I'll do that on recent one ... and you should not get a permission denied for an working ip
1122223444 M * maharaja btw - do you maintain any changelog for the vserver rcX ?
1122223466 M * Bertl http://linux-vserver.org/ChangeLog26
1122223480 M * maharaja mhm
1122223486 M * maharaja maybe its related to the port?
1122223496 M * maharaja ill try to start bind as root
1122223505 M * Bertl well, 53 is fine ...
1122223521 M * Bertl unless something else already bound there ...
1122223538 M * maharaja mhm, its working
1122223542 M * maharaja bind -u bind does not work
1122223547 M * maharaja uhm
1122223549 M * maharaja "named -u bind"
1122223560 M * maharaja named when run as root seems to work
1122223569 M * maharaja i've got capabilities disabled and removed it from the ./configure line
1122223584 M * Bertl ah, well, that explains it
1122223590 M * maharaja mhm
1122223599 M * maharaja do you know how to "fix" that?
1122223604 M * Bertl bind is dropping all priviledges with cap disabled
1122223625 M * Bertl (when you try to run it as user != root)
1122223645 M * maharaja does seem logical to me
1122223679 M * Bertl well, if you _are_ rewriting the bind anyway, why not just remove the silly CAP_RESOURCE stuff
1122223690 M * maharaja does a chroot fix the security problems related to starting bind as root?
1122223706 M * Bertl this would allow bind to run with capabilities inside a guest without raising the capset
1122223714 M * maharaja i'm not rewriting it, im doing some ugly hacks to get it working
1122223716 M * maharaja ;)
1122223727 M * Bertl then do the right ones :)
1122223731 M * maharaja hehe
1122223741 M * Bertl IIRC there are several existing patches to make bind work as expected
1122223741 M * maharaja not quite feeling to be able to do that
1122223763 M * Hollow Bertl: could you take a look at http://home.xnull.de/work/vserver/libvserver/util/syscall-errors.txt
1122223772 M * sannes what has changed from 2.0rc4 - rc8.1 ?
1122223782 M * Hollow wrt to implement own error strings in the lib
1122223796 M * sannes nevermind
1122223877 M * Bertl Hollow: hmm, is this already a mapping, or the current state?
1122223891 M * Hollow this is what i found out by looking at the sources ;)
1122223908 M * sannes in the changelog, what does " added new network interface" mean?
1122223931 M * Hollow the error code and what it is meant to say
1122223970 M * Bertl sannes: two new syscall commands to ease the life of userspace (add/rem network ip)
1122224026 M * Bertl Hollow: okay, will go through the list and comment ...
1122224088 M * maharaja Bertl: do you think a chrooted bind run as root is safe enough?
1122224111 M * maharaja Bertl: or do you know any possibility to alter the ports reserved for the root user?
1122224141 M * Bertl maharaja: depends on the criterion you apply to safety
1122224151 M * Bertl maharaja: no, lowports are root only
1122224378 M * maharaja ty bertl
1122225757 M * Hollow Bertl: the txt is completed now
1122225769 M * Hollow at least with things i found out ;)
1122225788 M * Bertl k
1122225811 M * Bertl I guess that might lead to error code consolidations ...
1122225824 M * Hollow probably..
1122226049 M * Hollow Bertl: is it possible to define own error codes in the kernel?
1122226062 M * Bertl yes, but it's a PITA
1122226070 M * Hollow *g*
1122226705 J * daniel_hozac ~daniel@c-6f1472d5.010-230-73746f22.cust.bredbandsbolaget.se
1122226716 M * Bertl wb daniel_hozac!
1122226724 M * daniel_hozac thanks.
1122233975 Q * sannes Read error: Connection reset by peer
1122236823 Q * Loki|muh Read error: Connection reset by peer
1122237159 Q * war- Read error: Connection reset by peer
1122237532 J * Loki|muh loki@satanix.de
1122238109 M * Bertl Hollow: you around?
1122238123 M * Hollow yep
1122238152 M * Bertl I have a preliminary version (for the syscall) wanna test with gcc4?
1122238167 M * Hollow sure
1122238221 M * Bertl http://vserver.13thfloor.at/Experimental/SYSCALL/syscall_shiny5.h
1122238301 M * Hollow ../../lib/syscall.c: In function 'vserver':
1122238301 M * Hollow ../../lib/syscall.c:30: error: bp cannot be used in asm here
1122238635 M * Bertl try to remove "ebp" from the __sysc_regs line
1122238657 M * Bertl (in the i386 section :)
1122238873 M * Hollow Bertl: works
1122238878 M * Hollow i.e. compiles
1122238923 M * Hollow and works too ;)
1122239298 M * Bertl excellent ...
1122241870 J * Aiken ~james@tooax6-088.dialup.optusnet.com.au
1122241954 N * bipsen bipsen_Zz
1122242169 M * Bertl welcome Aiken! night bipsen_Zz!
1122242276 M * Aiken good morning
1122242388 M * Bertl Aiken: what non x86 archs do you have available?
1122242411 M * Aiken the alpha and the sparc IPX (40Mhz)
1122242435 M * Bertl ah, sparc, great! could you test compile something there? (i.e. is it linux?)
1122242455 M * Aiken yes
1122242462 M * Aiken on both counts
1122242479 M * Bertl but I guess it's not running a vserver kernel (yet), right?
1122242485 M * Aiken not yet
1122242518 M * Bertl hmm .. hmm ... you have a config for that machine? (2.6 kernel config)?
1122242838 M * Aiken it is a 2.2.something at the moment
1122242854 M * Aiken when that image was built the 2.4 kernel were no good for sparc32
1122242862 M * Aiken still have to try a 2.6 kernel
1122242865 M * Bertl i.c. ...
1122242931 M * Bertl okay, if you like to test a 2.6 kernel, and you go through the config stuff, I can cross compile the kernel for you ...
1122243046 M * Aiken ok
1122243188 Q * yarihm Quit: Leaving
1122244130 M * Aiken what would be the easiest way of sending you the .config?
1122244376 M * Bertl whatever you prefer (dcc, email, url)
1122244711 M * Aiken emailed it, easier
1122244773 M * Aiken Fatal error: Image too large to fit in destination
1122244774 M * Aiken Error loading /boot/vmlinux-2.6.12.3-vs2.0-rc8.1
1122244791 M * Aiken that is what I get with the kernel I just cross compiled :(
1122244825 M * Aiken 2.8 meg in size
1122244875 M * Bertl hmm ... so it requires a 'small' kernel?
1122244894 M * Bertl does it support gzipped images?
1122244938 M * Aiken don't know
1122244992 M * Bertl http://www.obsolyte.com/sun_ipx/ <- something like this one?
1122245068 M * Aiken it is 10Mhz faster than mine
1122245070 M * Aiken that is it
1122245082 M * Aiken I have always just used vmlinux
1122245096 M * Bertl okay, give me a minute .. I'll check something
1122245153 M * Aiken considering how slow the cpu is they are not a bad little machine.
1122245309 M * Bertl yeah, you can do a lot with 40Mhz risc power ...
1122245311 M * Aiken gzipping the image did not good,
1122245398 M * Aiken the kernel is 2.8 meg and the 2.2.26 kernel the machine is running is 1.8meg
1122245553 M * Bertl we can move a lot of stuff out of the kernel I guess ...
1122245601 M * Bertl a lot of things look like kitchensink to me ... in your config :)
1122245697 M * Aiken already removed a lot from the default config
1122245748 M * Bertl 2623384 here without symbols
1122245781 M * Bertl do you need any of the crypto stuff?
1122245843 M * Bertl and what is your boot filesystem?
1122245868 M * Bertl is the mouse in use? or just via console/remote
1122245933 M * Aiken the crypto stuff should have been removed, all access is minicom for serial console 
1122246070 M * Bertl okay ...
1122246151 M * Bertl I guess either nfs or ext3 could go too?
1122246210 M * Bertl which stuff could be built as modules?
1122246216 M * Aiken it current has ext2 but want to eventually move it to ext3
1122246219 M * Aiken make a lot of use of nfs
1122246229 J * ang3l4u Angel@81.181.82.216
1122246236 M * Bertl welcome ang3l4u!
1122246327 M * Bertl 2369796 Jul 25 01:05 vmlinux
1122246328 M * ang3l4u heya Bertl
1122246378 M * Aiken smaller than I am getting
1122246389 M * Bertl 1070848 Jul 25 01:04 net/built-in.o
1122246408 M * Bertl is the largest part, if I could put it into modules, it would probably get below 2M
1122246494 M * Bertl do we need to get below 2M or do you want to try the 2.3M one?
1122246564 M * Aiken I am wondering if there is a 2 meg limit
1122246857 M * Aiken size is something I don't like with the current kernels
1122246896 M * Aiken upgrade the kernel and move more to modules to keep the core size down so the machine will still boot
1122247036 M * Aiken having the nvram in an IPX die is not fun
1122247145 M * Aiken seting up the nvram every power on until it was fixed
1122247205 M * Aiken it's booting
1122247251 M * Bertl ah?
1122247253 M * Aiken Linux localhost 2.6.12.3-vs2.0-rc8.1 #4 Mon Jul 25 09:12:02 EST 2005 sparc unknown unknown GNU/Linux
1122247259 M * Bertl hey great!
1122247268 M * Bertl congrats!
1122247272 M * Aiken 2485590 Jul 25 09:15 vmlinux-2.6.12.3-vs2.0-rc8.1
1122247295 M * Aiken make CROSS_COMPILE=sparc-sun4c-linux-gnu- ARCH=sparc image
1122247312 M * Aiken I found the image target and once I had it small enough that started booting
1122247334 M * Bertl perfect ... there is also a 'size' option in the kernel
1122247346 M * Bertl (which allows you to compile with minimum size)
1122247350 M * Bertl -Os basically
1122247466 M * Aiken well it has booted, now to get a usefull kernel config working
1122248072 M * Aiken I wonder if the size limit is about 2 1/2 meg, it seemed once I got below that a kernel finally booted
1122248261 M * Aiken not good http://pastebin.com/319924
1122248268 M * Aiken can i have a copy of the kernel you compiled to try
1122248338 M * Bertl sure
1122248406 M * Bertl http://vserver.13thfloor.at/Stuff/aiken-vmlinux-01 (not complete yet)
1122248469 M * Bertl okay, it's there, config too
1122248861 M * Aiken what was the test you wanted me to run?
1122248988 M * Aiken still grabbing your kernel and just had another of mine die :(
1122249473 Q * monrad Quit: Leaving