1162339203 N * _ Hollow
1162339417 J * gdm ~gdm@www.iteration.org
1162339771 M * Bertl wb gdm!
1162340179 J * duckx ~Duck@tox.dyndns.org
1162340408 J * Skram ~Mark@hermes.sentiensystems.com
1162340426 J * bronson ~bronson@66.160.177.216
1162340883 J * Greek0 ~greek0@85.255.145.201
1162340925 J * mugwump ~samv@watts.utsl.gen.nz
1162341112 M * Bertl anybody here who wants to figure the changes between 2.1.0 and 2.1.1? (have some ideas how to simplify that, but I'm not too eager to do it :)
1162341329 M * Rusty` Nov  1 03:30:14 apache pure-ftpd: (?@?) [ERROR] Unable to switch capabilities : Operation not permitted
1162341330 M * Rusty` :(
1162341352 M * Bertl with 2.1.x or 2.0.x ?
1162341376 M * Rusty` 2.1.x
1162341385 M * Rusty` the newest vserver patch :)
1162341389 M * Bertl interesting, could you strace -fF it?
1162341420 M * Bertl (I assume somebody of the pure-ftpd team had a 'good' idea :)
1162341497 Q * WorkRoey Ping timeout: 480 seconds
1162341514 M * Rusty` Bertl: http://195.38.113.36/strace
1162341567 M * Rusty` 16899 geteuid32()                       = 0
1162341567 M * Rusty` 16899 capset(0x19980330, 0, {CAP_CHOWN|CAP_DAC_READ_SEARCH|CAP_SETGID|CAP_SETUID|CAP_NET_BIND_SERVICE|CAP_NET_ADMIN|CAP_SYS_CHROOT|CAP_SYS_NICE, CAP_CHOWN|CAP_DAC_READ_SEARCH|CAP_SETGID|CAP_SETUID|CAP_NET_BIND_SERVICE|CAP_NET_ADMIN|CAP_SYS_CHROOT|CAP_SYS_NICE, }) = -1 EPERM (Operation not permitted)
1162341567 M * Rusty` 16899 rt_sigprocmask(SIG_BLOCK, ~[RTMIN], [], 8) = 0
1162341577 Q * bronson Quit: Ex-Chat
1162341586 M * Bertl yeah, not very 'secure' :)
1162341593 M * Rusty` what?
1162341603 M * Bertl but what I wonder is, this should not be blocked in devel
1162341633 Q * ntrs Remote host closed the connection
1162341653 M * Rusty` oh?
1162341660 M * Bertl let me double check this here ...
1162341684 M * Rusty` i don't understand, sry :)
1162341685 M * Bertl (will take a moment, have to get my test system into an usable state first)
1162341692 M * Rusty` ah so, ok
1162341712 J * ntrs ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com
1162341720 M * Bertl wb ntrs!
1162341720 J * s0undt3ch_ ~s0undt3ch@81.193.57.229
1162341807 J * bronson ~bronson@66.160.177.216
1162341930 J * WorkRoey ~katz@h-69-3-4-130.mclnva23.covad.net
1162342155 Q * s0undt3ch Ping timeout: 480 seconds
1162342155 N * s0undt3ch_ s0undt3ch
1162342363 Q * sladen Ping timeout: 480 seconds
1162342394 J * s0undt3ch_ ~s0undt3ch@81.193.57.136
1162342452 Q * mnemoc Ping timeout: 480 seconds
1162342537 J * sladen paul@starsky.19inch.net
1162342693 Q * s0undt3ch Ping timeout: 480 seconds
1162342693 N * s0undt3ch_ s0undt3ch
1162342920 Q * bonbons Quit: Leaving
1162343346 J * mnemoc ~amery@kilo105.server4you.de
1162343944 M * gdm Bertl: thanks, but i shouldn't have left!! guess it was some server prob or other (don't worry, not vserver ;-)
1162343952 A * gdm off to bed now anyway. goodnight!
1162343967 M * Bertl have a good one!
1162344151 J * FCOJ ~mordur@dsl-201-4.hive.is
1162344163 M * Bertl wb FCOJ!
1162344334 M * FCOJ Thanks
1162344511 J * ms_ ~ms@arkansas.doc.ic.ac.uk
1162344581 Q * FCOJ Quit: Leaving
1162345878 J * derjohn2 ~aj@dslb-084-058-252-133.pools.arcor-ip.net
1162346269 J * Borg- borg@cube.benet.uu3.net
1162347830 M * Bertl daniel_hozac: ping?
1162348022 Q * eyck Ping timeout: 480 seconds
1162348073 J * cazimir ~cazimir@p548A2F71.dip0.t-ipconnect.de
1162348079 M * Bertl welcome cazimir!
1162348090 M * cazimir huhu
1162348097 M * Skram Hiya, Bertl 
1162348100 M * Skram long time no talk
1162348107 M * Bertl indeed
1162348140 M * cazimir I run a Ubuntu host and want to build an Gentoo guest. howto? :>
1162348164 M * Bertl well, probably the best way is to get a template or isntalled gentoo
1162348176 M * Skram it is easiest just to go online and get an image/template
1162348199 M * cazimir an skeleton guest?
1162348203 M * Bertl cazimir: there are no tools (AFAIK) which create gentoo guests on debian/ubuntu
1162348239 M * Bertl cazimir: I assume you know how to install gentoo?
1162348252 M * cazimir yup
1162348272 M * cazimir but i got a big problem. i made a skeleton guest 
1162348293 M * Bertl so one way except for downloading a stage 3) would be to install it into a dir
1162348301 M * cazimir then extract the stage archive and tried to start the guest
1162348324 M * Bertl that's probably not so far off 
1162348331 M * cazimir then it says something with the init script
1162348349 M * Bertl yeah, because you probably didn't specify an init style
1162348361 M * Bertl (default is sysv, which is not what gentoo uses)
1162348371 M * Bertl try to change that to 'plain'
1162348372 M * cazimir how do i specify?
1162348390 M * Bertl either on the skeleton creation (see vserver - build --help)
1162348403 M * Bertl or by adding that to the config tree
1162348419 M * cazimir aahh ok thx alot :))
1162348467 M * Bertl Rusty`: still around?
1162348758 J * Aiken_ ~james@tooax6-108.dialup.optusnet.com.au
1162348807 Q * bronson Ping timeout: 480 seconds
1162349087 Q * Aiken Ping timeout: 480 seconds
1162349322 M * cazimir it works. thank you very much Bertl
1162349336 M * Bertl you're welcome!
1162349572 M * cehteh mhm
1162349770 Q * ensc Killed (NickServ (GHOST command used by ensc_))
1162349780 J * ensc ~irc-ensc@p54B4D6A5.dip.t-dialin.net
1162350260 J * cazimir_ ~cazimir@p548A2AA7.dip0.t-ipconnect.de
1162350727 Q * cazimir Ping timeout: 480 seconds
1162352756 Q * mejo Quit: Lost terminal
1162358483 M * daniel_hozac Bertl: pong
1162358487 Q * adrien Ping timeout: 480 seconds
1162358496 M * Bertl daniel_hozac: good morning :)
1162358538 M * Bertl daniel_hozac: I wonder, what caps do we set with devel kernels nowadays?
1162358565 M * daniel_hozac hmm?
1162358566 M * Bertl and is there a difference between cap mask and actual caps?
1162358630 M * Bertl what I mean is, for example Rusty` observed issues with pure-ftp, which 'lowers' caps 
1162358676 M * Bertl now, the devel branch cap system is designed to allow 'normal' cap manipulations despite of the fact that the actual caps are limited by the capability mask
1162358708 M * Bertl nevertheless, pure-ftp seems to hit a cap(set) limit ... so the question now is
1162358727 M * Bertl is that a bug in the kernel, or in userspace?
1162359360 M * daniel_hozac i can't reproduce it here.
1162359381 M * daniel_hozac maybe the process had already dropped privs prior to calling that (or it had none to start with)?
1162359857 Q * ms_ Ping timeout: 480 seconds
1162360148 M * Bertl daniel_hozac: could be .. don't know either
1162360179 M * Bertl we'll have to ask Rusty` once he's back
1162360282 M * Bertl daniel_hozac: so just for my understanding, what caps (from the capset PoV) does a guest with 0.30.211 (on 2.1.x) have?
1162360302 M * daniel_hozac 0xfffffeff is what capget is returning, at least.
1162360327 M * daniel_hozac so IMHO it should be possible to set all of those.
1162360345 M * Bertl except for 0x100, sounds fine to me
1162360376 M * daniel_hozac right.
1162360465 M * Bertl how does a process get CAP_SETPCAP?
1162360579 M * Bertl hmm, seems I'm not the first one wondering about that :)
1162360581 M * Bertl http://lkml.org/lkml/1999/12/5/138
1162360629 M * Bertl http://lwn.net/Articles/199932/
1162360636 M * daniel_hozac hehe
1162360796 M * Bertl so assumed Rusty` didn't hit some bug, that should be fine
1162360822 M * daniel_hozac yep.
1162360930 M * daniel_hozac does http://linux-vserver.org/index.php?title=Capabilities_and_Flags&diff=1855&oldid=1850 look sane?
1162360972 M * daniel_hozac (i.e. did i do the math right? it's still pretty early... ;))
1162361011 M * Bertl looks good ...
1162361116 M * Bertl it looks really good actually, i.e. we should change the kernel side defs accordingly
1162361228 M * daniel_hozac hehe, why? isn't the bit number more useful?
1162361253 M * daniel_hozac (unless you're reading the values)
1162361267 Q * FireEgl helium.oftc.net neutron.oftc.net
1162361267 Q * Piet helium.oftc.net neutron.oftc.net
1162361267 Q * bj helium.oftc.net neutron.oftc.net
1162361267 Q * fosco helium.oftc.net neutron.oftc.net
1162361267 Q * Wonka helium.oftc.net neutron.oftc.net
1162361267 Q * trippeh helium.oftc.net neutron.oftc.net
1162361326 J * FireEgl FireEgl@Sebastian.Atlantica.US
1162361327 J * Piet hiddenserv@tor.noreply.org
1162361327 J * fosco fosco@konoha.devnullteam.org
1162361327 J * trippeh atomt@x.vx.no
1162361327 J * bj ~bj@insanefactory.com
1162361327 J * Wonka produziert@chaos.in-kiel.de
1162361369 M * Bertl yeah, well, every notation has it's advantages and disadvantages
1162361387 M * Bertl but we might add them as comments for example
1162361471 M * daniel_hozac yeah
1162362161 Q * Piet Remote host closed the connection
1162362371 J * Piet hiddenserv@tor.noreply.org
1162362437 M * Bertl increadible ... just found another task/pid related bug
1162363500 J * Nigromante ~n@197.Red-80-35-167.staticIP.rima-tde.net
1162363508 M * Bertl welcome Nigromante!
1162363526 M * Nigromante hi
1162363677 P * Nigromante Abandonando
1162363980 T * Bertl http://linux-vserver.org/ <- new and shiny | latest stable 2.02.1, exp 2.02.2-rc4, devel 2.1.0, exp 2.1.1-rc46, stable+grsec 2.0.2.1 | util-vserver-0.30.211 | libvserver-1.0.2 & vserver-utils-1.0.3 | 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 ;)
1162363985 M * Bertl *rc46
1162364245 Q * Piet Remote host closed the connection
1162364303 M * Bertl okay folks! I'm off to bed for now ... have a good one everyone! cya!
1162364310 N * Bertl Bertl_zZ
1162364420 J * Piet hiddenserv@tor.noreply.org
1162364897 Q * mugwump Ping timeout: 480 seconds
1162365635 J * Aiken__ ~james@tooax8-223.dialup.optusnet.com.au
1162365726 J * meandtheshell ~markus@85-124-233-91.work.xdsl-line.inode.at
1162365739 J * mugwump ~samv@watts.utsl.gen.nz
1162366037 Q * Aiken_ Ping timeout: 480 seconds
1162366594 J * ms_ ~ms@arkansas.doc.ic.ac.uk
1162367709 J * dna_ ~naucki@145-246-dsl.kielnet.net
1162372621 J * bonbons ~bonbons@83.222.36.111
1162373177 J * bronson ~bronson@c-71-198-75-160.hsd1.ca.comcast.net
1162374275 Q * shedi Quit: Leaving
1162374522 Q * node Ping timeout: 480 seconds
1162374805 J * eyck eyck@ghost.anime.pl
1162375105 J * node ~dwindsor@stanford.columbia.tresys.com
1162375590 Q * ruskie Read error: Connection reset by peer
1162375821 J * ruskie ~ruskie@ruskie.user.oftc.net
1162376176 J * shedi ~siggi@130.208.221.254
1162377704 J * s0undt3ch_ ~s0undt3ch@81.193.63.219
1162377874 Q * _cob Remote host closed the connection
1162378097 J * ||Cobra|| ~cob@146.50.22.204
1162378112 Q * s0undt3ch Ping timeout: 480 seconds
1162378112 N * s0undt3ch_ s0undt3ch
1162378419 Q * cazimir_ Remote host closed the connection
1162379224 J * Godsey ~jason@proto.OpenSYSV.com
1162380263 M * h01ger its not possible to use 127.0.0.1 in different vservers at the same time, is it?
1162380284 M * harry no
1162380289 M * harry not afaik at least :
1162380290 M * harry :)
1162380388 M * daniel_hozac of course it is.
1162380399 M * daniel_hozac it's rewritten to the guest's first IP address on bind and connect.
1162380411 M * harry ah...
1162380422 M * h01ger oh nice. thats new
1162380431 M * harry right, i knew that!!!
1162380439 M * harry tnx for reminding me, daniel_hozac :)
1162380448 M * daniel_hozac no, that's not new. that's how it has worked for a really long time.
1162380462 M * harry rewrite source ip option ???
1162380467 M * harry that's quite new iirc
1162380495 M * harry anyway... ==> shopping!!
1162380502 M * harry (for 0xf00d
1162380516 M * daniel_hozac the rewrite source IP option is new, yes, but the destination address has been rewritten forever.
1162380579 M * harry ha!... hmm... nice :)
1162380584 M * harry vserver rules!
1162380594 M * harry (stating the obvious is sometimes good for selfesteem ;))
1162381603 J * mejo ~jonas@dslb-084-058-125-037.pools.arcor-ip.net
1162381611 M * mejo hello
1162381621 M * mejo i still didn't manage to compile a debian kernel with vserver support.
1162381631 M * mejo i simply cannot find the vserver options in make menuconfig
1162381673 M * mejo anyway the debian package linux-source-2.6.18 claims to have the vserver patch vs2.0.2.2-rc2 applied.
1162381997 Q * ms_ Ping timeout: 480 seconds
1162382237 Q * shedi Quit: Leaving
1162382352 M * h01ger mejo, there are even debian binary packages with vserver support
1162382610 M * mejo h01ger: i know, but i need a selfcompiled host kernel.
1162382622 M * mejo h01ger: make-kpkg --subarch vserver is what i was searching for
1162382694 M * Hollow daniel_hozac: do you know what happens with mounted filesystems if the last namespace using it vanishes? does it get umounted automagically or will it be in an incosistent state (i.e. fsck or sth like that is needed)
1162382733 Q * Aiken__ Quit: Leaving
1162382743 M * daniel_hozac it should get unmounted when the last reference is dropped.
1162382748 M * Hollow good
1162383063 M * derjohn daniel_hozac, wasnt it you who fixed "nfs mount inside a guest"? but how does it work? Do I put a line in  /etc/vservers/<xid>/fstab ? or s.th. special ?
1162383092 M * daniel_hozac the guest's fstab is executed in the host's context.
1162383129 M * daniel_hozac NFS mounts inside a guest is for running mount -t nfs ... as root in the guest.
1162383138 M * daniel_hozac (which requires secure_mount and binary_mount ccaps)
1162383186 M * daniel_hozac but for NFS mounts, fstab.remote would be better as that's limited to the guest's IP address(es).
1162384106 M * derjohn daniel_hozac, I i give secure_mount and binary_mount: That wont be limited to a special mount only, i.e. the user can mount everything in range? Is it possible to limit the mount the "type nfs"? (Which  would make sense IMHO)
1162384125 M * daniel_hozac no.
1162384152 M * daniel_hozac either the guest is able to mount stuff, or it's not :)
1162384322 M * derjohn If I put the mountline in the /etc/vservers/.../fstab: does it differ from mountind a dir into /var/lib/vservers/.../my/mount/point ? Background: Since newer RCs I have the problem that apache2 says: Warning: DocumentRoot [ /var/lib/vservers/.../my/mount/point ] does not exist. But I sweat it's there and I can edit files there with vi ...
1162384329 M * Hollow daniel_hozac: http://paste.linux-vserver.org/604 :))
1162384334 M * Hollow noe everything works :)
1162384336 M * Hollow *now
1162384369 M * daniel_hozac Hollow: awesome!
1162384393 M * daniel_hozac derjohn: as user apache?
1162384396 M * Hollow (even with openvz ;)
1162384419 M * derjohn daniel_hozac, good Q. I try !
1162384468 M * daniel_hozac and you really ought to put in fstab.remote, otherwise the mount request will come from the host's address, and the subsequent requests from the guest.
1162384548 M * derjohn daniel_hozac, did "su www-data" and was able to edit /var/lib/vservers/.../my/mount/point/bla/file
1162384581 M * derjohn (yes, inside the guest)
1162384631 M * daniel_hozac is it there when you're starting the guest?
1162385018 M * derjohn daniel_hozac, yes, I mounted it permanently from the hosts fstab
1162385052 J * rgl ~Rui@84.90.9.212
1162385055 M * rgl hello
1162385162 M * daniel_hozac hi
1162385183 M * daniel_hozac derjohn: very strange...
1162385183 M * phedny is it possible to give a guest access to an IP address, only for connect() and not for listen()?
1162385195 M * daniel_hozac phedny: SNAT?
1162385227 M * derjohn daniel_hozac, with older rc it worked (well older linux in general - I just replace a host)
1162385266 M * daniel_hozac derjohn: tried strace'ing it to see what's failing?
1162385289 M * phedny daniel_hozac: I can't come up with a good solution with SNAT atm
1162385297 M * derjohn hmn I did a "stat dir" ... I 'll try a strace with grep ...
1162385330 M * phedny daniel_hozac: I want to give my guest a non-routable IP for communication with other guest (for incoming messages)
1162385335 M * derjohn shit, strace isnt grep-able ...
1162385344 M * phedny daniel_hozac: but it needs to be able to do DNS and sometimes connect to an HTTP server
1162385350 M * rgl how can I pass arguments to init?
1162385355 M * daniel_hozac derjohn: -fF -o file
1162385356 M * phedny daniel_hozac: but I don't want to allocate a dedicated IP for only these tasks
1162385375 M * daniel_hozac phedny: why can't you SNAT those connections from the guest
1162385487 M * phedny daniel_hozac: would that work? as the 172.-IP I bound to the lo interface
1162385571 M * phedny and SNAT is in POSTROUTING
1162385572 M * daniel_hozac rgl: add the command with arguments to apps/init/cmd.start separated by newlines?
1162385587 M * daniel_hozac phedny: yes, so?
1162385595 M * daniel_hozac phedny: should work fine IMHO.
1162385610 M * phedny hmm, I'll just try on my test machine
1162385644 M * derjohn daniel_hozac, stat64("/var/www/...", {st_mode=S_IFDIR|0755, st_size=18, ...}) = 0
1162385652 M * derjohn daniel_hozac, it's a32 bit guest in a 64 env
1162385656 M * derjohn *host
1162385673 M * derjohn hm, with the ipv6 patch added
1162385682 M * daniel_hozac derjohn: so it looks fine, no?
1162385695 M * derjohn daniel_hozac, define: fine :)
1162385695 M * daniel_hozac there has to be some error in that trace.
1162385705 M * daniel_hozac stat returned 0?
1162385709 M * rgl daniel_hozac, ah thx :)
1162385711 M * derjohn stat64 !
1162385717 M * daniel_hozac yep, so?
1162385720 M * derjohn even if the guest is 32 bit ?
1162385722 M * daniel_hozac yes.
1162385727 M * daniel_hozac especially if it's 32-bit.
1162385734 M * rgl daniel_hozac, the cmd.prepare is run before init in the context of host?
1162385771 M * rgl humm or in the context of guest?
1162385940 M * daniel_hozac cmd.prepare is run to set the runlevel and such by default.
1162385962 M * Hollow daniel_hozac: what do you think about http://paste.linux-vserver.org/605 ?
1162385971 M * daniel_hozac you have the entire scripts directory to run things before and after the init has been started.
1162385993 M * daniel_hozac Hollow: hmm, point?
1162385996 M * Hollow this should fix the race conditions with chmod in redhat/initpost (and soon gentoo)
1162386016 M * daniel_hozac ... but require chmod to be present in the guest.
1162386023 M * Hollow uhm..
1162386024 M * Hollow ;)
1162386034 M * rgl daniel_hozac, I see. thx :D
1162386087 M * rgl is there a way to redirect the output to a file? or better to host syslog?
1162386107 M * Hollow daniel_hozac: ok ,so what about making a stripped down chmod where only numeric values can be set? (i.e. no ugo+rwx style)
1162386114 M * daniel_hozac have a fifo in apps/init/tty or so.
1162386126 M * daniel_hozac Hollow: yeah, i was thinking about that yesterday.
1162386169 M * rgl daniel_hozac, humm, do you have an example?   I'll have to use mkfifo apps/init/tty then how do I connect the fifo to syslog?
1162386177 M * Hollow daniel_hozac: mind if i prepare something?
1162386194 M * daniel_hozac Hollow: not at all, thanks.
1162386208 M * daniel_hozac rgl: have a daemon read from it and forward it to syslog? :)
1162386246 M * rgl daniel_hozac, ah I though there was a "easier" way :)
1162386281 M * daniel_hozac not that i know of.
1162387008 M * Hollow daniel_hozac: http://paste.linux-vserver.org/606
1162387278 M * daniel_hozac Hollow: should probably use the conversion functions already present.
1162387303 M * Hollow ?
1162387319 M * Hollow ah
1162387320 M * Hollow ensc_fmt
1162387353 M * daniel_hozac isNumberUnsigned
1162387421 M * Hollow ok.. mess
1162387422 M * Hollow :P
1162387424 M * Hollow brb
1162387571 M * daniel_hozac well, i gotta go home, i'll bbl.
1162387695 J * shedi ~siggi@dsl-149-109-85.hive.is
1162388094 M * Hollow daniel_hozac: meh.. you can't supply base to isNumberUnsigned...
1162388126 M * Hollow the only solution would be sth like chroot-sh 0x644 /foo/bar/file
1162388414 J * virtuoso_ ~s0t0na@shisha.spb.ru
1162388498 Q * virtuoso Read error: Connection reset by peer
1162388503 N * virtuoso_ virtuoso
1162388726 J * ms_ ~ms@arkansas.doc.ic.ac.uk
1162388923 M * rgl how do I add /dev/console into a guest?
1162388934 M * Hollow you don't
1162388967 M * Hollow it's insecure to have /dev/console inside
1162388997 M * rgl sure, but its for testing why upstart (ubuntu new init replacement) is not running
1162389010 M * rgl so isn't there an insecure way to have it there?
1162389035 M * Hollow not that i know of..
1162389053 M * Hollow it will most notably disable your keyboard, and kill your wm session (if any)
1162389471 P * mejo 
1162389927 M * sladen Bertl_zZ: got a moment to pop into #upstart on Feenode ?
1162390678 J * jet ~jim@AC9E54FC.ipt.aol.com
1162390953 J * mejo ~jonas@dslb-084-058-125-037.pools.arcor-ip.net
1162391327 M * ms_ sladen: um, are you Paul Sladen as of the RISC OS community?
1162391377 M * sladen ms_: I can Paul Sladen, but I haven't really done any RISC OS stuff, although I am a Brit
1162391400 M * ms_ mmm. ok. I must be confused then.
1162391459 M * sladen the top result on google for "paul sladen" "risc os" is me though
1162391489 M * sladen however the following post alludes to there having been another one (or two?) at Wakefield RISC OS shows
1162391605 M * ms_ oh! I know you! You're nottingham lug?
1162391616 M * ms_ andy davidson, HE etc etc?
1162391640 M * ms_ [pt]otty mailing list etc?
1162391710 M * ms_ sheesh. small world
1162391914 Q * shedi Quit: Leaving
1162391915 J * sebastian ~sebastian@pD957DDE3.dip.t-dialin.net
1162392166 M * sladen ms_: yup, and yourself?
1162392188 M * sladen ah, mister Sackman.  I'm in London btw (and have been for the last 3)
1162392198 M * ms_ sladen: Matthew Sackman, used to be in Notts, HE etc, now PhD at Imperial
1162392204 M * sladen ms_: does mister will-h still 0wn your ass?
1162392210 M * ms_ he never did!
1162392214 M * ms_ that was alistar
1162392217 M * sladen s/ass/login/
1162392223 M * ms_ oh, well sort of.
1162392232 M * sladen *grin*
1162392240 M * ms_ dept.of.Computing are sort of outside his realm
1162392245 M * ms_ we do our own stuff really.
1162392262 M * sladen As it happens, I did some work for Andy on Friday
1162392275 M * ms_ oh right. is he still doing that printing co?
1162392290 M * ms_ (I've been out of touch for some time)
1162392473 M * FaUl mh
1162392671 J * Rusty`` ~rusty@catv-50623279.catv.broadband.hu
1162392674 M * Rusty`` hello
1162392748 M * FaUl hey Rusty`` 
1162392799 M * FaUl mh, i get mails from my crond which does a cronjob with writing the output of vserver-stat into a file periodically, it says: vc_get_task_xid(26897): No such process 
1162392805 M * FaUl any idea why it does this?
1162392967 M * sladen ms_: no, heading up Ebuyer
1162393008 J * adrien ~mirc@ppp08-89.dsl.citenet.net
1162393011 M * ms_ sladen: eek. That's a change then. So what are you up to these days then?
1162393013 M * adrien Hi
1162393017 M * adrien I'm new to verser
1162393030 M * adrien I'm having hard time to create my first guest
1162393042 M * sladen ms_: Freelance various.  And time spent on Ubuntu
1162393049 M * rgl hey adrien 
1162393062 M * adrien I want to perfom a fc5 installation
1162393082 M * adrien but a regular vserver test build -m yum -- -dfc5
1162393087 M * adrien is too slow
1162393101 M * adrien I want to use my local web server
1162393104 M * adrien to provide the rpms
1162393114 M * adrien but I have no Idea where I can change this
1162393122 M * adrien in the .distributions/fc5
1162393131 M * adrien I only have a setup for apt nothing for yum
1162393305 M * rgl adrien, maybe this helps, /usr/lib/util-vserver/distributions/fc5/yum.repos.d ?
1162393342 J * mejo_ ~jonas@dslb-084-058-152-197.pools.arcor-ip.net
1162393345 M * rgl adrien, I think you should do a cp -R /usr/lib/util-vserver/distributions/fc5 /etc/vservers/.distributions/ and change it there
1162393363 M * adrien ok I understand now
1162393366 M * adrien thanks !
1162393370 M * rgl adrien, I mean change inside /etc/vservers/.distributions/fc5/yum.repos.d/
1162393386 M * adrien so if vserver doesn't find /etc/vservers/.distributions/fc5/yum.repos.d/
1162393394 M * adrien it will fail-back to  /usr/lib/util-vserver/distributions/fc5 
1162393401 M * rgl yup, tahts it :D
1162393432 M * adrien bye the way, waht would you recommand to make a centos guest
1162393445 M * adrien the yum method, the apt, the rpms ?
1162393451 M * rgl I used this: YUM="yum -d2" vserver example build \
1162393451 M * rgl 	-m yum \
1162393451 M * rgl 	--context 10 \
1162393451 M * rgl 	--hostname example.com \
1162393451 M * rgl 	--interface eth0:192.168.1.11/24 \
1162393453 M * rgl 	--initstyle sysv \
1162393454 M * rgl 	-- -d centos4
1162393461 M * rgl ops.. what a paste!  sorry for the spam
1162393494 M * rgl be sure to always give the guest an unique context id (in this case it was 10)
1162393747 Q * mejo Ping timeout: 480 seconds
1162393747 N * mejo_ mejo
1162393901 M * daniel_hozac Hollow: 0x is hexadecimal, 644 looks like octal to me ;)
1162393953 M * daniel_hozac Hollow: and 0644 should work fine.
1162394026 M * Hollow daniel_hozac: right.. octal..
1162394123 M * Hollow daniel_hozac: here we go: http://paste.linux-vserver.org/609
1162394154 M * Hollow some messages not correct yet..
1162394186 M * Hollow c&p annoyance
1162394243 Q * bronson Ping timeout: 480 seconds
1162394314 M * daniel_hozac Hollow: help message is missing the final \n too.
1162394323 M * Hollow yup, already fixed ;)
1162394338 M * daniel_hozac other than that, looks fine, you'll commit it?
1162394389 M * Hollow yup
1162395243 M * Hollow daniel_hozac: done.
1162395249 M * Hollow with some fixes ;)
1162395511 Q * jet Ping timeout: 480 seconds
1162395749 M * daniel_hozac ok ;)
1162395754 M * daniel_hozac thanks.
1162395841 Q * michal` Ping timeout: 480 seconds
1162396083 M * Hollow daniel_hozac: what about adding a temporary gentoo-experimental init style? this way people could test the new baselayout...
1162396107 M * Hollow i guess it will take >= 6 months until it gets stable
1162396110 M * daniel_hozac any reason we can't use gentoo for that?
1162396128 M * daniel_hozac gentoo is already discouraged in the docs, right?
1162396136 M * Hollow we could do that... but we should check for the right baselayout version then IMO
1162396162 M * daniel_hozac sure.
1162396192 M * Hollow ok, will take a look... i already made such a check for the gentoo/initpost
1162396194 M * FaUl vc_get_task_xid(29518): No such process 
1162396197 M * FaUl wtf/
1162396243 M * daniel_hozac FaUl: vserver-stat parses ps's output to get the values, if the process dies before ps is done outputting things and vserver-stat has parsed it, you'll get that.
1162396257 M * daniel_hozac (that's my subtle way of saying, don't use vserver-stat ;))
1162396317 M * FaUl daniel_hozac: ok, so i'll add another 2>/dev/null
1162396319 M * FaUl ;-)
1162396414 M * adrien is there a debug/verbose option for "vserver build", because it's not very explicit on what it's doing ?
1162396454 M * daniel_hozac vserver --debug ... should show you the commands, i think.
1162396463 J * michal` ~michal@81.169.139.228
1162396476 M * adrien ok I'll try
1162396625 M * daniel_hozac Rusty``: can you still reproduce the issues you were having yesterday?
1162396632 M * daniel_hozac Rusty``: do you have a complete trace available?
1162396673 M * Rusty`` ooo
1162396698 M * Rusty`` strace: http://195.38.113.36/strace
1162396705 M * Rusty`` is'nt working solution yet :(
1162396733 M * goblin daniel_hozac, wow, you must really like working on vserver... you're always so useful
1162396749 M * goblin daniel_hozac, are you paid for it, or do you do it totally for pleasure?
1162396765 M * goblin I was always wondering about people like you... it's sort of amazing :-)
1162396818 M * daniel_hozac goblin: i'm not paid for it ;)
1162396850 M * goblin you must have plenty of spare time then... :-)
1162396879 M * daniel_hozac well, outbound SSH isn't blocked at school ;)
1162396911 M * goblin aah :-) nice ;-)
1162396943 M * goblin yeah, school time was nice...
1162397090 M * daniel_hozac Rusty``: just to make sure, what does uname -a say?
1162397127 M * daniel_hozac Rusty``: and what distro is the guest running?
1162397144 J * bronson ~bronson@c-71-198-75-160.hsd1.ca.comcast.net
1162397237 M * daniel_hozac Hollow: hmm, how do i resolve a conflict with svn? i've fixed the file, but svn st still lists it as C.
1162397260 M * Hollow svn resolved <file>
1162397262 M * Rusty`` rusty@apollo:~/pote$ uname -a
1162397262 M * Rusty`` Linux apollo 2.6.17.8-grsec2.1.9-vs2.0.2-rc28 #4 SMP Thu Oct 26 20:12:38 CEST 2006 i686 GNU/Linux
1162397269 M * daniel_hozac ah, thanks.
1162397274 M * Rusty`` the guest and the host distro is debian too
1162397278 M * daniel_hozac Rusty``: well, uh, that's expected then.
1162397282 M * daniel_hozac you're running stable.
1162397309 M * daniel_hozac only devel has the capability masking features.
1162397323 M * Rusty`` oh
1162397352 M * Rusty`` devel is production-ready? :)
1162397418 M * daniel_hozac well, it's called devel for a reason :) i believe there are people running it in production though.
1162397511 M * daniel_hozac also note that AFAIK there are no grsec merges with devel
1162397756 M * Rusty`` :(
1162397818 N * Bertl_zZ Bertl
1162397823 M * Bertl morning folks!
1162397825 M * daniel_hozac morning Bertl!
1162397849 M * daniel_hozac oo, while i remember it, should binary_mount be required for mounting tmpfs?
1162397862 M * Bertl ah, so Rusty``s issues are resolved
1162397897 M * Bertl tmpfs, hmm, no, not really
1162397921 M * daniel_hozac the vfs_kern_mount fix requires that, i guess we should check for tmpfs magic as well?
1162397932 M * Hollow :)
1162397946 M * Rusty`` morning Bertl ).
1162397947 M * Rusty`` :)
1162397963 M * Hollow is it possible to also check for devpts and skip NODEV?
1162398084 M * Bertl a lot of stuff is _possible_ :)
1162398102 M * Hollow ;)
1162398230 M * daniel_hozac the NODEV is added long before we know the magic though.
1162398415 M * daniel_hozac does BME let you dev-ify an existing mount point with mount --bind /path/a /path/b? i.e. will /path/b have accessible device nodes if /path/a is mounted NODEV?
1162398443 M * Bertl don't think so
1162398538 M * Bertl hmm, shouldn't a check for 'data' be much more appropriate in vfs_kern_mount()?
1162398550 M * Bertl (for the binary mount case)
1162398578 M * daniel_hozac i'm rather curious about the original capable check that was there...
1162398598 M * daniel_hozac and it seems BME does allow that.
1162398630 M * Bertl really? hmm, why doesn't it apply nodev?
1162398655 M * daniel_hozac well, the new mount point doesn't have nodev, so isn't that kind of expected?
1162398695 M * Bertl it's early in the morning, but I thought be applied nodev basically unconditional?
1162398709 M * mejo does the vserver kernel patch introduce options that i should see in menuconfig?
1162398715 M * daniel_hozac well, if !capable(CAP_SYS_ADMIN)
1162398718 M * daniel_hozac i tested on the host.
1162398754 M * daniel_hozac (i was thinking about moving the NODEV to later in the process, and i was trying to find out if it mattered to more than new mounts)
1162398767 M * daniel_hozac mejo: yes. an entire submenu.
1162398777 M * mejo on the first level?
1162398785 M * Hollow daniel_hozac: http://paste.linux-vserver.org/610 ;)
1162398809 M * daniel_hozac mejo: yes.
1162398860 M * daniel_hozac Hollow: looks really nice.
1162398970 M * Hollow daniel_hozac: :) now just the init style check on start is missing, will upload the patch then..
1162398983 M * daniel_hozac ok, sounds good.
1162399218 M * nox Hollow: something like apt-cacher would be nice for gentoo guest. Just compiling every package only one time and providing as stage3 for the other guest
1162399297 M * daniel_hozac emerge with binpkgs should do that, no?
1162399309 M * daniel_hozac assuming you have the same settings on your guests and host.
1162399331 M * daniel_hozac Hollow: btw, the fstab modifications should probably be conditionalized on the existance of the directories.
1162399342 M * Hollow nox: see gentoo vserver howto for shared binpkgs
1162399356 M * Hollow daniel_hozac: the directories are created it initpost too
1162399378 M * Hollow or do you mean on the host?
1162399479 M * daniel_hozac on the host.
1162399492 M * Hollow yup.. will check for that..
1162399496 M * daniel_hozac i.e. if i just want one Gentoo guest, i don't have much to share with :)
1162399599 M * daniel_hozac any progress on util-vserver-izing the vemerge, etc. from the ebuild?
1162399663 M * nox doesn´t a bind,rw mount is dangerous if you have not trustworthy guests?
1162399703 M * nox but rest seems 2 be really nice for me (still) nongentoo user
1162399716 M * Hollow daniel_hozac: not yet.. but it's on my todo
1162400273 J * shedi ~siggi@inferno.lhi.is
1162400859 M * Bertl daniel_hozac: I think we should check the fstype instead of other stuff
1162400871 M * Bertl daniel_hozac: will prepare a patch after dinner ...
1162400884 N * Bertl Bertl_oO
1162400945 M * Hollow [17:31] <Bertl> it's early in the morning => [18:07] <Bertl> ... after dinner ...
1162400948 M * Hollow :P
1162400978 M * Hollow s/dinner/breakfast/
1162401110 M * blizz lol
1162401283 A * Skram waves to #vserver 
1162401329 J * frag24|maxh ~1_jolly@84.137.123.197
1162401333 M * frag24|maxh Hi
1162401355 M * frag24|maxh Is there anybody who may help me?
1162401389 M * Skram Hi
1162401413 M * frag24|maxh We are currently using a vserver-version that is pretty old (only one configfile in /etc/vservers/). My question is if I can update it to a newer version with an 2.6er Kernel
1162401423 Q * ruskie Quit: Disconnecting from stoned server.
1162401432 M * frag24|maxh I'm asking because the configfiles changed and stuff like that
1162401448 M * Skram you could always just migrate to the new config file structure..
1162401456 M * frag24|maxh I can't afford to miss vservers for a long period of time
1162401463 M * frag24|maxh ok
1162401466 M * frag24|maxh how do I do that?
1162401468 M * Skram look in the kernel.. i forget if there was an option to use legacy configs
1162401469 M * Skram i forget
1162401487 M * Skram frag24|maxh: look at the documentation for the "new" way of config files and do that...
1162401504 M * frag24|maxh can't I just keep the vserverdirectories and just wirte configfiles with the new system for them
1162401512 M * frag24|maxh or are they incompactible?
1162401519 M * Skram I am not sure
1162401524 M * frag24|maxh too bad
1162401528 M * frag24|maxh thanks anyways
1162401558 M * Skram there are 113 others in here, I bet someone will get back to 'ya
1162401559 Q * nebuchadnezzar Quit: ERC Version 5.1.4 (IRC client for Emacs)
1162401563 M * nox well guests doesn´t need vserver specific stuff
1162401586 M * nox so i c no problem with that
1162401616 J * ruskie ~ruskie@ruskie.user.oftc.net
1162401635 M * nox just take your time to study the flowerpage
1162401665 J * nebuchadnezzar ~nebu@zion.asgardr.info
1162401667 M * frag24|maxh thanks a lot for your help. If this is the way you do it... don't save it under support on the website
1162401679 M * daniel_hozac frag24|maxh: there's a script to convert it.
1162401689 M * frag24|maxh I didn't find something looking like an updatedescription
1162401691 M * frag24|maxh ok cool thanks!
1162401692 M * daniel_hozac frag24|maxh: old configs are also still supported.
1162401702 M * frag24|maxh So it's only the Kernel and the Configfiles?
1162401706 M * daniel_hozac (with the correct options in the kernel)
1162401714 M * frag24|maxh Anything else can be like it was right?
1162401725 M * daniel_hozac well, it depends on what utils you were using and how you configure the kernel.
1162401742 M * daniel_hozac ancient utils should still work, given that you enable the legacy version ID.
1162401769 M * daniel_hozac but, obviously, in order to get all the new goodies the 2.6 based kernels have, you'll need more recent utils, and a more recent config.
1162401856 M * frag24|maxh I hope the damn things will still work afterwards :-) Otherwise I'M screwed
1162401862 M * frag24|maxh but thanks a lot for your kind help!!!
1162401901 M * nox frag24|maxh: 2 be sure test them on a testserver
1162401930 M * frag24|maxh I got another server running with the new vservers
1162401935 M * frag24|maxh they are working fine
1162402019 Q * frag24|maxh Quit: Adieu! • :: ««« (Gamers.IRC) »»» www.gamersirc.net ::
1162402079 Q * bronson Ping timeout: 480 seconds
1162402840 J * lilalinux ~plasma@dslb-084-058-211-182.pools.arcor-ip.net
1162402845 N * Bertl_oO Bertl
1162402849 M * Bertl back now
1162402860 M * daniel_hozac wb!
1162402915 M * Bertl Hollow: don't you eat breakfast before you go to sleep? :)
1162402933 M * Hollow :P
1162402981 M * Bertl daniel_hozac: okay, I think I have a plan to improve the mount flags and capabilities, but I want to get 2.1.1 out asap
1162402998 M * Bertl daniel_hozac: is there any change we absolutely need in 2.1.1?
1162403020 M * Bertl (if not, I'd suggest to leave that for 2.1.2 and make sure that this is released in the next month or so
1162403035 M * daniel_hozac Bertl: nah, i guess most people are content with giving binary_mount as well if they want guests mounting tmpfs ;)
1162403036 M * Bertl (i.e. I do not want to ahve such long cycles in devel as we had now)
1162403074 M * Bertl daniel_hozac: we can simply add the tmpfs magic to the check there, if that helps
1162403103 M * Bertl but I don't want to mess with the various checks which IMHO have grown pretty ugly
1162403115 M * daniel_hozac i guess you should be asking Hollow, he was the one who ran in to it ;)
1162403126 M * Bertl Hollow: *?
1162403128 A * Hollow is confused
1162403141 M * Hollow shouldn't 2.1.1 be released as 2.2.0 later on?
1162403152 M * Bertl nah, but the 2.1.1 base
1162403162 M * Bertl look, my plans look like this:
1162403171 M * Bertl - releas devel 2.1.1 (NOW!)
1162403193 M * Bertl - declare 2.1.1 as 2.2.0-rc1 and start fixing cleaning ...
1162403214 M * Bertl - continue 2.1.x (maybe already call it 2.3.x?)
1162403239 M * Hollow i'd prefer 2.3.x then...
1162403254 M * Bertl thing is, I want some things like the lo stuff to go into 2.2.0
1162403259 M * daniel_hozac yeah, me too.
1162403267 M * daniel_hozac i really think the network stuff should be in there.
1162403269 M * Bertl so the early 2.2.0-rcx will get some changes
1162403271 M * daniel_hozac (i.e. IPv6 as well)
1162403298 M * Bertl another option would be:
1162403304 M * Bertl - releas devel 2.1.1 (NOW!)
1162403320 M * Bertl - start gathering all what's required for 2.2.x into 2.1.2+
1162403338 M * Bertl - declare 2.2.2+ as 2.2.0-rc*
1162403347 M * Bertl *2.1.2+
1162403356 M * daniel_hozac i'd prefer the latter.
1162403360 M * Hollow indeed
1162403379 M * Bertl okay, then we do it like that, after all a 2.0.3 release will be due
1162403425 M * daniel_hozac sounds good.
1162403439 M * Bertl so, Hollow, do you need the TMPFS check in 2.1.1?
1162403443 M * Hollow no
1162403460 M * Bertl okay, then we don't touch it, and rewrite the entire mnt check system in 2.1.2
1162403497 M * Bertl things in my queue for 2.1.2 are:
1162403503 M * Bertl - loopback isolation
1162403515 M * Bertl - is_interrupt()
1162403529 M * Bertl - (re)mount handling
1162403546 M * Bertl - context quota (either finish or rip out)
1162403570 M * Bertl - revisit nfs tagging
1162403602 M * Bertl ipv6 will start immediately after 2.1.1 release, but I'm not sure it will be 2.1.2 stuff
1162404058 M * Hollow sounds like a plan :)
1162404107 M * Hollow daniel_hozac: http://people.linux-vserver.org/~hollow/util-vserver/delta-gentoo-feat01.diff
1162404110 M * SNy "I'll put on my slightly larger glasses."
1162404116 M * SNy d;
1162404151 M * Hollow hehehe
1162404166 M * Hollow "who comes down here? what are you talking about?"
1162404186 M * SNy "Why are you giving me the secret sign to shut up?"
1162404190 M * Hollow :)
1162404193 M * SNy "Brilliant."
1162404220 M * daniel_hozac $_AWK instead of awk? and why [[ rather than test or [?
1162404241 M * Hollow because [[ is built-in => faster, also does it handle empty LHS
1162404258 M * daniel_hozac test/[ is also built-in, no?
1162404259 M * Hollow awk, yep..
1162404268 M * Hollow it is?
1162404272 M * daniel_hozac i thought so.
1162404283 M * Hollow but afaik it does not handle empty LHS very well
1162404289 M * daniel_hozac and are quotes really such a pain in the ass? ;)
1162404298 M * Hollow yep
1162404299 M * Hollow :)
1162404320 M * Hollow but well... to fit in util-vserver ugly bash syntax we can change it of course :P
1162404347 A * Hollow hides
1162404356 M * daniel_hozac :P
1162404365 M * daniel_hozac will the mkdir calls need || :?
1162404382 M * daniel_hozac or is set -e not active for the build methods?
1162404395 M * Hollow uhm.. good question
1162404436 M * Hollow set -e is active
1162404468 M * daniel_hozac the inittab fixing is kinda ugly, IMHO. won't the echo mess up formatting and such?
1162404515 M * Hollow indeed..
1162404519 M * Hollow bad echo
1162404561 M * daniel_hozac can't you just pipe it along?
1162404562 M * Hollow i first had it without that ugly logic, but once you said "chmod may not be available inside guests" i thought: sed won't be too then :P
1162404581 M * daniel_hozac hehe.
1162404587 M * daniel_hozac and i guess that ought to be $_SED as well?
1162404604 M * Hollow yeah... i'm not used to these vars.. *bah*
1162404605 M * Hollow :)
1162404616 M * Hollow i expect a working $PATH ;)
1162404669 M * Hollow we should even use $_CAT then
1162404676 M * Hollow and $_ECHO?
1162404677 M * Hollow ;)
1162404749 M * Bertl okay, so I start to write a changelog now :)
1162404759 M * Hollow hurray!
1162404829 M * daniel_hozac not $_CAT, you're using $_CHROOT_SH
1162404835 M * daniel_hozac echo is built-in :P
1162404848 M * daniel_hozac Bertl: great!
1162404872 M * daniel_hozac Hollow: and maybe fix the (c)s?
1162404903 M * Hollow you mean copyright?
1162404928 M * daniel_hozac yeah.
1162404940 M * Hollow yeah, why not..
1162404986 M * Hollow daniel_hozac: btw, [ and test are built-in and [[ is a compound command, whatever the difference is..
1162405057 M * daniel_hozac hehe, ok.
1162405316 M * daniel_hozac test is used rather consistently throughout the utils, so i guess using that would be best.
1162405362 M * daniel_hozac and isn't the fstab massaging suffering from the same problems as inittab?
1162405378 M * Hollow daniel_hozac: ok.. the only solution i see for the inittab is to use temporary file..
1162405384 M * Hollow yep..
1162405390 M * daniel_hozac yeah, that's the downside :(
1162405505 M * Hollow because simply pipe everything in one line would result in an empty inittab
1162405537 M * daniel_hozac yeah.
1162405568 M * daniel_hozac an inline sed might work, but then we have the chroot-thing.
1162405580 M * daniel_hozac well, would work.
1162405602 M * Hollow work yes, chroot no... that's why i wanted to use chroot-sh exec
1162405621 M * Hollow or simply use chroot
1162405678 M * daniel_hozac hmm, and any particular reason you're not using $SETUP_INITSTYLE?
1162405707 M * daniel_hozac or hmm, no, that won't be exported.
1162405714 M * Hollow nod
1162405937 Q * derjohn2 Ping timeout: 480 seconds
1162405967 J * derjohn2 ~aj@dslb-084-058-222-029.pools.arcor-ip.net
1162406257 M * Bertl wb derjohn2!
1162407163 J * Kerberos ~satan@85.138.138.2
1162407175 M * Bertl welcome Kerberos!
1162407181 M * Kerberos hello
1162407397 N * Zaki[] Zaki
1162407899 P * Kerberos isnt it obvious?
1162407994 J * kerberos ~satan@85.138.138.2
1162408010 M * Bertl wb kerberos! testing your client?
1162408046 M * kerberos bot?
1162408053 M * kerberos im new here..lol..
1162408060 M * Bertl well, I'm too :)
1162408102 M * daniel_hozac new being a relative term, or what? :P
1162408109 M * Hollow *giggle*
1162408117 M * Hollow daniel_hozac: here we go again: http://people.linux-vserver.org/~hollow/util-vserver/delta-gentoo-feat01.diff
1162408120 M * Bertl yeah, compared to the alps for example :)
1162408120 M * kerberos second time in this irc...
1162408148 M * Hollow uhm.. forgot copyright, but else it should be ok now..
1162408152 M * Bertl kerberos: well, I already gave you a warm welcome, no?
1162408161 M * kerberos sure..
1162408162 M * Bertl kerberos: so do you need anything?
1162408185 M * kerberos if i need..lol.. i need some docs.. to learn this stuff..
1162408197 M * Bertl linux-vserver.org?
1162408205 M * kerberos yeah started today..
1162408226 M * Bertl so, if you want to know something, just ask (see topic)
1162408296 M * kerberos well i got a exame tomorrow..
1162408300 M * daniel_hozac Hollow: still using [[ in vserver.functions, and missing quotes around $basemin/$basemaj in initpost, $_RM -f $inittabtmp? does $vdir make sense for popd?
1162408317 M * Bertl kerberos: about linux-vserver? :)
1162408329 M * kerberos nah numerical analisys..
1162408364 M * daniel_hozac Hollow: and still the old copyright ;)
1162408374 M * Bertl kerberos: interesting area, isn't it?
1162408407 M * Hollow daniel_hozac: how could i forget teh quotes!
1162408407 M * kerberos just a discipline on my course.. but offcourse its interesting..
1162408425 M * kerberos if it werent then what would i be doing on school...
1162408439 M * Bertl right
1162408493 M * kerberos theres documentation.. about vservers.. but creating one..puting the files in there..
1162408505 M * kerberos i didnt see much.. for my distro.. slackware
1162408526 M * kerberos i made one.. but.. apache wouldnt start...
1162408536 M * Bertl it's quite simple actually, if you have a working installation, that's fine
1162408544 M * kerberos and dmesg showed me some kernel bugs..
1162408564 M * Bertl that's unusual, could you upload them to paste.linux-vserver.org please?
1162408584 M * kerberos :X ...sory..
1162408592 M * Bertl apache not starting is typically the result of havin an apache already running on the host
1162408592 J * bronson ~bronson@66.160.177.201
1162408598 M * kerberos deleted everything..meanwhile..
1162408604 M * kerberos patch-2.6.18.1-vs2.1.1-rc44
1162408631 M * Bertl probably those were just warnings, btw, we are at rc46 now
1162408640 M * kerberos well i disabled.. apache on the host..
1162408670 M * Bertl well, if you want to investigate your 'issues', just let me know
1162408676 M * kerberos but tomorrow or something.. ill get into to it..
1162408680 M * kerberos ok thanks..
1162408685 M * kerberos need to study...
1162408689 M * Bertl you're welcome! have fun!
1162408711 M * kerberos just one thing..
1162408718 M * Bertl sure?
1162408727 M * kerberos should i use he 2.1.1 or stable
1162408735 M * Bertl depends on what you want
1162408747 M * Hollow daniel_hozac: ok, reload again... ;)
1162408747 M * Bertl stable is _rock_solid_stable_
1162408775 M * Bertl devel is devel, so it's usually quite stable, but might break here and there
1162408800 M * Bertl the rcs are still in flux, as well as the experimental releases
1162408844 M * kerberos i read somewhere that for starters it would be good to use 2.1.1
1162408851 M * kerberos but.. it something fails. on it..
1162408861 M * kerberos i would be going in circles..
1162408871 M * Bertl if soemthing fails, you should contact us, as it is considered a bug(tm)
1162408898 M * daniel_hozac Hollow: looks fine to me.
1162408919 M * Bertl kerberos: and usually (real) bugs are fixed within an hour or two
1162408919 M * Hollow good, will commit it then..
1162408930 M * daniel_hozac ok.
1162408942 M * kerberos yeah.. but if i dont create a "standard" vserver...
1162408947 M * Bertl kerberos: in 99% of all cases, Linux-VServer is not the cause for issues (i.e. it is a bad configuration or other issue)
1162408963 M * Bertl kerberos: there is nothing like a "standard" vserver
1162409011 M * kerberos so.. i can use other distros diferent from my own?
1162409033 M * Bertl kerberos: we are not providing a product for the masses ... Linux-VServer is a set of building blocks (given, with quite powerful tools) to allow you to get the most out of virtualization/isolation
1162409070 M * Bertl kerberos: you can basically run _any_ distro, within any other distro
1162409088 M * Bertl (as long as those distros can work with the Linux-VServer kernel)
1162409236 M * kerberos well i downloaded some slackware images..
1162409243 M * kerberos to see how they build them..
1162409311 M * kerberos i really need to study for my exame tomorrow...later..thanks..
1162409322 M * Bertl k, cya!
1162409420 M * Hollow Bertl: btw, did you see http://paste.linux-vserver.org/602 ? it is working again without init, and no baselayout-vserver anymore..
1162409451 M * Bertl wow, great! good work!
1162409474 M * Bertl I think folks will love it ...
1162409524 M * Hollow yep.. this will obsolete quite some stuff, also building guests will be a lot easier with 0.30.212: http://paste.linux-vserver.org/610
1162409563 M * Bertl ah, nice, utilizing the tar build stuff
1162409576 M * Hollow yep, no more gentoo specific scripts..
1162409583 M * Bertl when will 212 hit the masses?
1162409585 M * Hollow except those in util-vserver svn now ;)
1162409837 M * Hollow Bertl: well, i think for the gentoo part we might have to do some minor cleanups still... will test things the next days extensivly... for the rest you have to ask daniel_hozac
1162409866 M * Bertl daniel_hozac: *?
1162409950 M * daniel_hozac well, there isn't much that has changed from 0.30.211 yet, and i was hoping ncontext/nattribute and such would be around before 0.30.212.
1162409976 M * Bertl what about the vlogin issues?
1162409995 M * Hollow hm?
1162410013 M * daniel_hozac yes, i'd like to be able to investigate those further before then as well.
1162410014 M * Bertl vlogin is definitely broken in 311 for most archs, no?
1162410028 M * Hollow interesting..
1162410042 M * daniel_hozac works fine on x86 and x86_64, and i'm inclined to blame dietlibc for any issues.
1162410060 M * daniel_hozac since we don't do much in vlogin.
1162410067 M * Bertl no problem with the blaming, but ut makes it unuseable :)
1162410072 M * Bertl *it
1162410082 M * daniel_hozac yeah, it's going to be configurable in 212 for sure.
1162410082 M * Hollow ah... good old diet..
1162410094 M * Hollow ;)
1162410106 M * daniel_hozac would it be possible to try with a glibc version just to make sure?
1162410109 M * Bertl daniel_hozac: but if you need some debug info/testing, just let me know, I have a few archs which won't work :)
1162410119 M * daniel_hozac i don't think i have immediate access to any of the problematic archs.
1162410162 M * Bertl np, starting a glibc compile as we speak ...
1162410170 M * Hollow on ppc?
1162410174 M * Bertl yup
1162410296 M * daniel_hozac it could be that we assume things which are only true on x86(_64), but i don't think there are too many arch-specific things about pseudo-ttys, are there?
1162410297 M * Hollow any error message from vlogin?
1162410324 M * daniel_hozac no, but there's no output or anything until EOF.
1162410328 M * daniel_hozac (IIRC)
1162410389 M * daniel_hozac and as i recall, select() didn't return until EOF.
1162410393 M * Bertl yeah, you can type, but nothing shows up until you press CTRL-D
1162410469 M * daniel_hozac i guess it's a dietlibc __NR_select vs. __NR_newselect thing.
1162410492 M * daniel_hozac (though it IMHO seems strange to have a disfunctional API)
1162410500 M * daniel_hozac for the kernel, i mean.
1162410860 M * Bertl I can confirm, works fine with glibc on ppc
1162410878 M * Bertl (at least I assume it uses vlogin by default)
1162410891 M * daniel_hozac yep, it should.
1162410911 M * Bertl will check again with dietlibc, just to make sure
1162410930 M * Bertl any tests/traces you want ome to do?
1162410938 M * Bertl s/ome/me/
1162410957 M * daniel_hozac well, we tested before, didn't we? or am i misremembering this entirely and confusing it with another problem?
1162410998 M * daniel_hozac i don't know if waldi submitted a patch already for dietlibc to use _newselect instead.
1162411000 M * Bertl I remember some discussion with doener .. something like select vs newselect, yes, but I'm not sure that was related
1162411022 M * Bertl is the patch somewhere, because if so, I could test that here, no?
1162411118 M * doener hm, I don't even know newselect, so I guess that was someone else (or my memory is broken)
1162411138 M * Bertl doener: nah, in case of doubt, it's mine :)
1162411157 M * Bertl s/mine/my memory/
1162411192 M * daniel_hozac http://people.linux-vserver.org/~dhozac/p/m/delta-select-test01.diff maybe.
1162411420 M * Hollow http://www.genpets.com
1162411424 M * Hollow o.o
1162411573 M * Bertl just finished crawling through the 2.1.0->2.1.1 changes
1162411596 M * Bertl here is the collection (not in usable format yet), please have a look if I missed soemthing
1162411600 M * Bertl http://paste.linux-vserver.org/611
1162411796 M * Hollow wow.
1162411803 M * Hollow didn't realize it was that much :)
1162411971 M * Hollow Bertl: regarding INFO_INIT.. isn't it needed for fakeinit?
1162411991 M * Bertl yes?
1162412002 M * Hollow why is it legacy then?
1162412042 M * Bertl hmm, where?
1162412058 M * Hollow - legacy VXF_HIDE_NETIF|VXF_INFO_INIT
1162412073 M * Bertl ah, that means, that we add those for legacy now :)
1162412084 M * Bertl (to make it more backwards compatible)
1162412104 M * Hollow eh? so it is legacy?
1162412145 M * Bertl nah, thing is, old kernels (legacy ones) did use HIDE_NETIF and INFO_INIT by default
1162412166 M * Bertl now, when we handle legacy kernel commands, we didn't add those
1162412171 M * Bertl (but 2.1.1 does)
1162412185 M * Hollow ah ok...
1162412466 M * Hollow Bertl: sysfs tagging?
1162412495 M * Bertl yeah, I wondered myself, but we tag it :)
1162412513 M * Hollow heh
1162412566 M * Hollow ah nice, VCMD_get_vci... must have missed that too
1162412687 M * Hollow i miss VCMD_rlimit_stat
1162412715 M * Hollow and reset_rminmax
1162412725 M * Bertl ah, yes, must have missed those
1162412802 M * Hollow looks complete then from my PoV :)
1162412828 M * Borg- howdy..
1162412839 M * Bertl howdy Borg-!
1162412845 M * Borg- I just got struck by mount: permission denied.. 
1162412858 M * Borg- strace show it dies here:
1162412860 M * Borg- mount("192.168.65.128:/home/repository", "nfs", "nfs", 0xc0ed0000, 0x80589e0) = -1 EPERM (Operation not permitted)
1162412868 M * Borg- so generaly im not allowed to use mount on guest ?
1162412878 M * Bertl depends on the caps your guest has
1162412890 M * Borg- caps!.. aha.. :)
1162412923 M * Bertl http://linux-vserver.org/index.php?title=Capabilities_and_Flags
1162412961 M * Borg- yeah :) reading..
1162412970 M * Borg- wonder what cap I need for mount.. 
1162412982 M * Borg- probably CAP_SYS_ADMIN
1162412984 M * Bertl in your case, it's a network mount (nfs)
1162413005 M * Bertl nah, SYS_ADMIN is dangerous, guest root will own your host
1162413023 M * Borg- hmm.
1162413027 M * Bertl try secure_mount and binary mounts
1162413056 M * Hollow we should probably add a BIG FAT WARNING above bcaps ;)
1162413222 M * Bertl yes, something like: Note: adding any bcap to your guest _will_ harm security -- don#t do it unless you absolutely know what you're doing
1162413237 M * Bertl s/harm/reduce/
1162413353 M * Borg- fast question: I can modify ccaps on running context?
1162413368 M * Bertl yes
1162413474 M * Hollow Bertl: done..
1162413565 M * Bertl okay :)
1162413686 M * Hollow Bertl: btw, did you see the page about chroot breakouts and the barrier?
1162413696 M * Bertl not yet
1162413700 M * Hollow i gathered all known (at least to me ;) exploits...
1162413708 M * Hollow http://linux-vserver.org/Secure_chroot_Barrier
1162413752 M * Bertl yeah, search already found it, looks noce
1162413806 M * Bertl except for the misleading 'Exploit seems to work', IIRC
1162413837 M * Bertl but maybe that was already fixed in this version?
1162413846 M * Hollow well, just c&p :) but yeah, it is probably also shown if it didnÄt work
1162413883 M * Hollow another btw: http://wiki.linux-vserver.org/Documentation preliminary doc structure, maybe you can comment on this as well
1162413905 M * Hollow parts of the step-by-step guide have been transfered into it too
1162414402 M * Borg- oki doki :) im closer now :>
1162414409 M * Borg- mount works.. thx to secure_mount
1162414414 M * Borg- but now server refuses me ;>
1162414415 M * Borg- thx Bertl !
1162414470 M * daniel_hozac Bertl: any changes with the patch?
1162414537 M * Bertl I'm currently preparing a 0.30 dietlibc for ppc
1162415686 M * Borg- vattribute --set --xid 90 --bcap net_bind_service
1162415689 M * Borg- hmm this doesnt work..
1162415701 M * Hollow .oO(BIG FAT WARNING...)
1162415702 M * Hollow :)
1162415715 M * Borg- yeah yeah.. I need bloody nfs working, I 0wn all vservers anyway :>
1162415721 M * Borg- s/vservers/guests/
1162415736 M * Bertl did you try the suggested caps?
1162415751 M * Bertl and, why don#t you mount the nfs on guest startup?
1162415798 M * Borg- Bertl: yeah I tried.. but still it b0rks anout ports (cannot bind low port)..
1162415801 M * Borg- and RPC service..
1162415858 M * Borg- Bertl: does portmap need to be run both on client and server?
1162415864 M * Borg- I didnt used NFS from ages..
1162415877 M * Bertl you need a working/reachable portmap
1162415889 M * Bertl but once again, why do you want to mount it _inside_?
1162415916 M * Borg- I dont need really.. but I dont want to kill guest..
1162415923 M * Borg- at least im trying to avoid that
1162415927 M * Bertl so?
1162415996 M * Bertl use vnamespace to --bind mount stuff into your guest
1162416002 Q * rgl Ping timeout: 480 seconds
1162416035 Q * sp Ping timeout: 480 seconds
1162416076 M * Borg- lets try :>
1162416166 M * Borg- aha indeed.. much better results
1162416176 J * rgl ~Rui@84.90.9.212
1162416206 M * Borg- now I just get Permission Denied from Nfs server
1162416209 M * Borg- dunno why really
1162416246 M * Bertl probably the guest ip is not allowed to mount?
1162416266 M * Borg- hmm found the problem
1162416270 M * Bertl or you didn't mount with the guest ip?
1162416274 M * Borg- when I execute mount via vnamespace.. wrong IP is used
1162416286 M * Borg- IP from host is used not the guest one
1162416299 M * Borg- (info from tcpdump -p -i lo) :)
1162416330 M * Bertl yeah, you want to do a chbind too, I guess
1162416347 M * Borg- w/o first? :) chbind or vnamespace?
1162416371 M * Bertl doesn't matter
1162416486 M * Borg- mount: cannot mount block device 192.168.65.128:/home/repository read-only
1162416489 M * Borg- huh ;)
1162416525 M * Bertl daniel_hozac: upgrading to diet 0.30 + mandriva patches fixes it
1162416543 M * Hollow VEOF?
1162416561 M * daniel_hozac Bertl: ok, good.
1162416568 M * Bertl daniel_hozac: so I consider that solved, and it's definitely a dietlibc issue
1162416668 M * Bertl uploaded a working dietlibc SRPM to http://vserver.13thfloor.at/Stuff/MANDRAKE/
1162416705 M * Bertl (contains a select patch which might fix exactly that issue, not checked though)
1162416884 M * daniel_hozac looks like.
1162416980 M * Bertl was the feature matrix already migrated (new wiki)? 
1162416990 M * daniel_hozac Downloads
1162417003 M * daniel_hozac it's on+
1162417010 M * daniel_hozac hmm, was?
1162417031 M * Bertl http://linux-vserver.org/Installation_Considerations
1162417038 M * daniel_hozac yep.
1162417141 M * Bertl hmm, Hollow, we might need an even tighter table version
1162417170 M * Bertl at least on my browsers the feature matrix is almost two pages long?
1162417190 M * Hollow ah well.. it still has the old css class
1162417215 M * Hollow will clean that up tomorrow... still one css missing too
1162417221 M * Bertl okay, np
1162417238 M * Hollow there are probably other tables using the old class too
1162417275 M * daniel_hozac can't we make the tight version the default?
1162417306 M * Hollow yeah, probably a good idea
1162417317 M * Bertl derjohn2: ping?
1162417330 M * Hollow will also move it from MediaWiki:Common.css to the htdocs directly..
1162417346 M * derjohn Bertl, my brother is at home, /me is in office ;)
1162417365 M * Bertl ah, nice :)
1162417382 M * Bertl do you plan to update the changelog (with my help)?
1162417392 M * derjohn sure!
1162417418 M * derjohn but i have to add that i have to fecth my parents from airport in a couple of minute
1162417425 M * derjohn *minutres
1162417428 M * derjohn *minutes
1162417441 M * Bertl np, you'll be back later?
1162417442 M * derjohn but: Which Changelog do we plan to offer?
1162417469 M * Bertl http://paste.linux-vserver.org/611 (this is between 2.1.0 and 2.1.1)
1162417494 M * Bertl question is, do we break that down into devel patches?
1162417497 M * derjohn the last one ported to the wiki was devel rc28. there are more in the old wiki, I can port it.
1162417564 M * Hollow ok folks.. quite tired already... hope to be able to commit a 2.1.1 tomorrow morning ;)
1162417570 M * derjohn i was wrong. there is only rc27 in the oldwiki
1162417575 M * derjohn n8 Hollow !
1162417591 M * Hollow night!
1162417609 M * Borg- Bertl: I added entrey to fstab of guest..
1162417618 M * derjohn Bertl, for stable we simply offer one changelog 2.1.0 ->  2.1.1, for devel we have to offer rc28 - rc... 46 ?
1162417619 M * Borg- and still I get permision denied.
1162417636 M * derjohn Borg-, NFS ?
1162417646 M * Borg- derjohn: yes
1162417648 M * Hollow 2.1.0 and 2.1.1 are devel too ;)
1162417649 M * Bertl Borg-: you want to use fstab.remote
1162417665 M * Bertl Borg-: and I hope you have a recent kernel/patch
1162417671 M * derjohn or put it in /etc/vservers/<xid>/fstab
1162417674 M * Borg- 2.6.17.13
1162417679 M * Borg- no patches..
1162417685 M * derjohn Borg-, :)
1162417690 M * Borg- derjohn: it is there.. and doesnt work
1162417694 M * derjohn Borg-, vserver patchlevel ?
1162417706 M * Borg- mount: 192.168.65.128:/home/repository failed, reason given by server: Permission denied
1162417709 M * Borg- /etc/vservers/titanium0/fstab:4:1: failed to mount fstab-entry
1162417710 M * derjohn did you paste the line?
1162417711 M * Borg- [root@titanium-adm] uname -sr
1162417713 M * Borg- Linux 2.6.17.13-vs2.0.2.1
1162417730 M * Borg- 192.168.65.128:/home/repository /mnt/nfs        nfs     defaults
1162417733 M * derjohn Bertl, 2.0.2 ? NFS mount borken ?
1162417734 M * Bertl that will not allow you to do mounts inside
1162417754 M * derjohn Bertl, happy upgrading ;)
1162417768 M * Bertl Borg-: but it should work fine if you mount on the host
1162417777 M * Borg- Bertl: yeah.. I thought the same..
1162417783 M * Borg- ugly hack.. but screw it :>
1162417802 M * Borg- mount blabla /home/vservers/titanium1/mnt/nfs should work :> at least I hope it will
1162417807 M * Bertl just --bind mount it via fstab into place
1162417812 M * derjohn Borg-, do you run apache2 inside a guest?
1162417821 M * Borg- derjohn: nope.. just apache 1.3
1162417836 M * Borg- tomcat and some java stuff
1162417838 M * derjohn Borg-, I'm having problems with DocRoot on a NFS, looking for someone to confirm the prob,
1162417900 M * Borg- fuck.. sth is broken now.
1162417900 J * Aiken ~james@tooax6-074.dialup.optusnet.com.au
1162417908 M * Borg- killall5 doesnt works as it should
1162417917 M * Borg- my guest do not stop correctly
1162418052 M * Bertl hmm?
1162418072 M * Bertl welcome Aiken!
1162418078 M * Bertl Borg-: how does it fail?
1162418084 M * Aiken hi Bertl
1162418225 M * Borg- Bertl: http://paste.linux-vserver.org/612
1162418229 M * Borg- it never happended before :)
1162418235 M * Bertl Aiken: do I remember correctly, that you observed the vlogin issues too?
1162418279 M * Bertl Borg-: what's boa?
1162418289 M * daniel_hozac webserver?
1162418291 M * Borg- boa web server
1162418291 M * Aiken yes
1162418300 M * Borg- I can kill it from guest easly
1162418309 M * Borg- but killall5 from rc doesnt work
1162418316 M * Bertl strange ...
1162418319 M * Borg- killall5 works also when I login into guest
1162418335 M * Borg- its not only boa.. any process except inetd fails to kill
1162418343 M * Borg- tomcat too.. cvs..
1162418344 M * Aiken Bertl vlogin issue with alpha but it is fine for x86
1162418352 M * daniel_hozac Aiken: try applying http://people.linux-vserver.org/~dhozac/p/m/delta-select-test01.diff to dietlibc.
1162418381 M * Bertl Aiken: or adapt the mandrake version I jsut uploaded http://vserver.13thfloor.at/Stuff/MANDRAKE/dietlibc-0.30-1mdk.src.rpm
1162418402 M * daniel_hozac that just fixes it for ppc/ppc64 though ;)
1162418410 M * Bertl Borg-: what tool version?
1162418425 M * Aiken I'll give that patch a go
1162418426 M * daniel_hozac Borg-: do you have initscripts to stop those services?
1162418453 M * Bertl daniel_hozac: he is running a shutdown/kill script (see paste)
1162418492 M * daniel_hozac ah.
1162418539 M * Borg- sth like that
1162418585 M * Borg- that is my script: http://paste.linux-vserver.org/613
1162418589 M * Borg- it used to work just fine :>
1162418603 M * Borg- before I started to play w/ vattributed & stuff
1162418620 M * Borg- maybe I will just reboot entire host :>
1162418634 N * _mcp mcp
1162418855 M * Borg- hope it will boot up
1162418946 Q * dna_ Quit: Verlassend
1162419061 Q * bonbons Quit: Leaving
1162419237 Q * mugwump Ping timeout: 480 seconds
1162419283 J * mugwump ~samv@watts.utsl.gen.nz
1162419291 M * Bertl wb mugwump!
1162419382 M * mugwump thanks.  silly routing issues with upstream provider
1162419390 M * mugwump we were about to drop them as a peer ;)
1162419401 M * Bertl how's going?
1162419419 M * mugwump yeah, pretty good.  thrashing quite a bit tbh
1162419433 M * Borg- lalala ..
1162419439 M * Borg- system didnt came up fully
1162419441 Q * mejo Quit: leaving
1162419446 M * Borg- ping ok.. but I cannot access it
1162419450 M * Borg- sth is fucked up :>
1162419456 M * mugwump doing some interesting work with an application of P2P to git ... see gittorrent.utsl.gen.nz
1162419463 M * Borg- tumorrow my developers will bitch about that.. heh
1162419463 M * Bertl Borg-: strange .. but doesn't sound Linux-VServer related to me
1162419483 M * Borg- Bertl: yeah.. probably I b0rked sth today
1162419490 M * Borg- im too tired to go to work now..
1162419503 M * Bertl k, so we'll figure tomorrow ...
1162420066 M * Borg- good nite
1162420575 M * Aiken I am using dietlibc 0.29, daniel_hozac with that patch dietlibc compiles but util-vserver won't
1162420588 M * Aiken without that patch util-vserver compile and runs fine except for vlogin
1162420764 M * Aiken I'll try a different version shortly
1162420997 Q * eyck Ping timeout: 480 seconds
1162421080 Q * rgl Quit: Fui embora
1162421559 M * derjohn daniel_hozac, i send you a trace 
1162421646 M * Bertl derjohn: how is rc46 doing for you?
1162421702 Q * meandtheshell Quit: exit (0);
1162422441 M * derjohn Bertl, i am on rc45 :)
1162422452 M * derjohn rc46 cant be so long ago ...
1162422480 J * schMA ~markus@p5491D4DB.dip.t-dialin.net
1162422489 M * derjohn i checked http://paste.linux-vserver.org/611 and must say. ugh. ugh.
1162422572 M * Bertl welcome schMA!
1162422581 M * derjohn what was the change in rc46 ? daniel's CoW stuff ?
1162422598 M * Bertl that and some task disappearing issue :)
1162422613 M * derjohn so something essential...
1162422654 M * derjohn interesting: I had besides the NFS "DocRoot not found" a [tcb ... ] exhausted
1162422685 M * schMA one question: is there a vserver v2.0.2.1 patch for 2.6.16(.29) available?
1162422715 M * Bertl hmm, I don't think so, but let me check .. mind the question, why such an old kernel?
1162422766 M * schMA i'm using vserver on top of xen, and xen 3.0.3 has been released with 2.6.16.29...
1162422787 M * Bertl but for sure, they have more recent patches, no?
1162422805 M * Bertl IIRC, debian is combining linux-vserver and xen for 2.6.18+
1162422832 M * derjohn Bertl, yes, there are hybrid kernels with lastest vserver stable and xen
1162422850 M * Bertl schMA: latest version for 2.6.16 was http://vserver.13thfloor.at/Experimental/patch-2.6.16.20-vs2.0.2-rc22.diff
1162422866 M * schMA i don't think, they have for their stable release, but debian would be a good choice, i'm running on debian ;)
1162422884 M * derjohn Bertl, was that a typical rc45 issue: vxW: pid_task(5778,0): task ffff81013e1af040[#898,5766] did lookup ffff81010ebd27b0[#0,5778]
1162422884 M * derjohn vxW: !!! limit: ffff81011717c070[<NULL>,20] = 1 on exit
1162422890 M * schMA @Bertl: yep, that's the version, i'm currently running on
1162422895 M * derjohn or: could that be one ? 
1162423000 M * Bertl derjohn: the first one is a warning, which is fine
1162423013 M * Bertl derjohn: the second one is unusual ...
1162423029 M * Bertl derjohn: but probably fixed in rc46 :)
1162423054 M * Bertl schMA: ask waldi, he probably knows the details
1162423056 M * derjohn ok ... but still the Apache2's DocRoot on NFS issue
1162423104 M * Bertl derjohn: can't reproduce that here, but we can investigate that after 2.1.1 (nfs needs an overhault anyways)
1162423139 M * derjohn can't means: it works for you , or you dont have a setup that fits ?
1162423187 M * Bertl means, I don't have a setup where nfs and apache has issues
1162423199 M * derjohn apache2 ?
1162423216 M * schMA Bertl: ok, thank's a lot! I think, I will come back later to reach him online
1162423223 M * Bertl derjohn: nope, just 1.3
1162423247 M * Bertl schMA: okay, np, feel free to hang around though
1162424811 M * Bertl okay, nap attack here ... probably back later
1162424823 N * Bertl Bertl_zZ
1162425033 M * daniel_hozac derjohn: i really don't see what's failing here. stat succeeds, and returns a good mode, so it has to be some sort of bug in httpd/apr/whatever.
1162425108 M * daniel_hozac very odd that you get permission denied though when you try to access those files.
1162425119 M * daniel_hozac is that in the trace, btw?
1162425121 M * daniel_hozac i can't see it.