1215734418 M * sebastiandeutsch hello, i'm a complete vserver noob - i want to delete a totally deconfigured vserver, do i simply remove it from /etc/vservers and /path/to/vservers?
1215734449 M * darkpsi stop it first
1215734449 M * sebastiandeutsch k
1215734449 M * darkpsi or if you've already rm'd it, you'll have to vkill -c <context>
1215734459 M * darkpsi to ge the context number, vserver-stat
1215734460 M * Bertl fuse, for me, bears the mark of Erez Zadok :)
1215734465 M * darkpsi lol
1215734489 M * Bertl sebastiandeutsch: you best do 'vserver <name> delete'
1215734539 M * Bertl darkpsi: and that means, nice for prototyping, bad for production
1215734575 M * darkpsi agreed.
1215734630 M * darkpsi of course, the extra call time is an issue (indirection)  ... hardlinking has at most one "extra" layer, which is always present anyhow
1215734636 M * darkpsi so performance wise, you are correct
1215734641 M * sebastiandeutsch Bertl: thank you
1215734644 M * darkpsi but md5 summing is hardly uncorruptible
1215734647 M * darkpsi hash collision?
1215734673 M * Bertl I guess we can rule that out in real world scenarios
1215734685 M * darkpsi don't make me get out my binary powers calculator :)
1215734706 M * darkpsi it's also pretty slow
1215734714 M * darkpsi to rehashify
1215734721 M * darkpsi and i get a bad feeling doing it
1215734722 M * Bertl well, you don't do that very often, do you?
1215734734 M * daniel_hozac hash collisions are handled.
1215734751 M * Bertl really? wouldn't have done that :)
1215734755 M * darkpsi daniel:  really, how?
1215734772 M * daniel_hozac if two files have the same hash, the other attributes are verified.
1215734803 M * darkpsi other attributes...  
1215734819 M * darkpsi filename, filedate, permissions, those migth be intentionally different
1215734830 M * darkpsi filesize is safe i guess
1215734969 M * darkpsi am I about to be stabbed?
1215734971 M * darkpsi i'm trying to help
1215734986 M * darkpsi unioning, fi it can be made to work--fast--, IS the "right way"
1215735022 M * daniel_hozac just write a daemon that uses inotify if you can't be bothered to run vhashify once a week.
1215735029 M * daniel_hozac (or once a month, or whatever your requirements are)
1215735043 M * darkpsi inotify-simple enough
1215735052 M * darkpsi you're talking per file
1215735060 M * darkpsi it's an ugly bear that moves quickly :)
1215735150 M * darkpsi which attributes does hashify consider?  RTFM is an acceptible answer, if any :)
1215735188 M * Bertl probably all attributes (or most of them)
1215735268 M * darkpsi alright.  i'm gonna sulk about my aufs problems later.
1215735277 M * daniel_hozac mode, uid, gid and size.
1215735288 M * Bertl darkpsi: btw, did you do any performance tests yet?
1215735291 M * darkpsi any idea why bind grabs the machine by the ventricles?
1215735297 M * daniel_hozac mtime is configurable.
1215735326 N * Guest2 Genghis
1215735350 M * Bertl darkpsi: you didn't answer the subversion part, or did I just miss it?
1215735357 M * darkpsi oh, you're right. didn't
1215735363 M * darkpsi as for performance comparisons, no i have not
1215735366 N * Genghis Guest11
1215735368 M * darkpsi rest assured hardlinks wioll do better
1215735400 M * Bertl well, I'm sure about that too, but it would be interesting how much better :)
1215735412 M * darkpsi also, btw, i noticed numerous interesting options within aufs that had to do with inode numbering
1215735424 M * darkpsi it may have a worthwile configuration, and mayh just be closer than any of us think
1215735437 M * darkpsi like i said, if I find a way, you'll hear about it
1215735452 M * Bertl excllent
1215735546 M * darkpsi kernel => version 2, patchlevel 6, sublevel 20, extraversion NULL (Homicidal Dwarf Hampster) -- who names these things?
1215735564 M * darkpsi vserver => what's the best way to get the precise version? i have 2.2.0-r1
1215735577 M * Bertl Linus does :)
1215735604 M * Bertl well, 2.2.0-r1 ... is that a gentoo kernel version?
1215735633 M * darkpsi 2.2.0-r1 is the vserver patch number
1215735643 M * darkpsi i wonder what a psychologist would say about that (Linus's naming fetish)
1215735679 M * daniel_hozac no, that's Gentoo version number.
1215735680 M * Bertl if that is true, I'd suggest to upgrade to vs2.2.0.7 and try again
1215735702 M * darkpsi it's a stock kernel, not genkernel
1215735707 M * Bertl if it is gentoo, then look for a 2.6.22.19 kernel (with vserver)
1215735721 M * darkpsi unpacked, patched w/ vserver, patched with aufs, and compiled
1215735765 M * Bertl well, 2.2.0-rc1 was somewhere in april 2007 
1215735792 M * Bertl it was the first release candidate, and at least 21 followed (fixing bugs :)
1215735808 M * Bertl then the vs2.2.0 was 'released'
1215735831 M * Bertl since then, we did (at least) 7 fixup releases (with minor bug fixes/improvements)
1215735877 M * Bertl daniel_hozac: what was the check/permission iconsistancy you mentioned last time?
1215735889 M * daniel_hozac fs/attr.c:inode_change_ok
1215735937 M * darkpsi hold on a sec.
1215735962 M * darkpsi i've tried this twice, with identical verisons of all things involved, once on a gentoo x86_32, and once on a from-scratch x86_64
1215735964 M * daniel_hozac i think we should either move that check to fs/namei.c:permission, or remove it completely.
1215735968 J * doener ~doener@i577BAA72.versanet.de
1215735988 M * darkpsi the x86_32 (gentoo) works okay, with bind, the x86_64 does not (bind starts, but hangs the system when it's stopped)
1215735999 M * Bertl daniel_hozac: the vx_check or the barrier stuff?
1215736012 M * darkpsi i'm going to veify it right now, on the x86_32 (gentoo+vserver+aufs), making a machine, and installing bind (emerge) i
1215736017 M * darkpsi i'll let you know
1215736018 M * daniel_hozac the PROC_SUPER_MAGIC thing.
1215736035 M * darkpsi it's a gentoo system, but a stock kernel, btw
1215736047 M * daniel_hozac with a Gentoo patch.
1215736072 Q * doener_ Ping timeout: 480 seconds
1215736110 M * daniel_hozac Bertl: does a guest have access to any proc-entries that should only be readable?
1215736161 M * Bertl I'm more worried about the pts and barrier
1215736196 M * Bertl we could move that check up into the filesystems (except for the barrier stuff)
1215736206 M * daniel_hozac yeah...
1215736219 M * Bertl but I think that wouldn't really help here, would it?
1215736233 M * Bertl because IIRC, something was hitting this check, no?
1215736240 M * daniel_hozac well, we need some way to allow it for certain files.
1215736281 M * daniel_hozac it seems a lot is permitted right now...
1215736285 M * Bertl please could you try to explain the 'original' issue to me?
1215736325 M * Bertl wasn't it something related to truncate vs. write or so=
1215736333 M * daniel_hozac yes.
1215736376 M * daniel_hozac the problem is that things like sysctl and pam_loginuid.so and other things that want to write to /proc use O_TRUNC.
1215736397 M * Bertl okay
1215736429 M * daniel_hozac this takes us to do_truncate, which calls notify_change (which calls inode_change_ok for /proc).
1215736446 M * Bertl still I don't see why that affects inode attributes?
1215736452 M * daniel_hozac hmm?
1215736460 M * daniel_hozac oh, the truncate sets the size to 0.
1215736478 M * Bertl ah, okay, well, then the test is legitimate
1215736487 M * daniel_hozac not exactly...
1215736492 M * Bertl question is, why do we block that one, but not the write itself?
1215736500 M * daniel_hozac exactly.
1215736509 M * daniel_hozac we need it higher up in the chain.
1215736528 M * daniel_hozac (with some bit to disable it)
1215736543 M * Bertl now, another question is, why does sysctl mess with proc?
1215736555 M * daniel_hozac it writes to /proc/sys/...
1215736572 M * darkpsi hey daniel, the -r1 is is because i copied my .config from an gentoo kernel
1215736585 M * darkpsi make mrproper didn't clean it up, which doesn't sound too.. proper
1215736611 M * darkpsi as i stated before, i made it with a makefile.
1215736618 M * Bertl daniel_hozac: is/was sys_sysctl deprecated and replaced by /proc/sys?
1215736639 M * daniel_hozac yeah.
1215736645 M * darkpsi wget , untar, cp .config, make mrproper, make defconfig, make.. etc
1215736673 M * darkpsi (patching in there too)
1215736682 M * darkpsi no gentoo patch
1215736703 M * daniel_hozac and you don't know what versions you used?
1215736720 M * darkpsi i told you.
1215736743 M * darkpsi broke down kernel => and asked if there was an aurthoritative number on the vserver
1215736747 M * darkpsi i have 2.2.0
1215736753 M * daniel_hozac 2.6.20-vs2.2.0 is unlikely, as that never existed.
1215736754 M * darkpsi -r1 was spurious on my part
1215736779 M * darkpsi what do you mean that never existed?
1215736786 M * daniel_hozac 2.6.20.4-vs2.2.0 was the first 2.2.0.
1215736814 M * darkpsi okay, I'll grab that version of the kernel
1215736830 M * darkpsi but again
1215736834 M * daniel_hozac 2.6.20.20-vs2.2.0.4 is the most recent version for 2.6.20.
1215736836 M * darkpsi i doubt that's the prob.
1215736842 M * daniel_hozac (released in october last year...)
1215736860 M * darkpsi the prob is bind squeezing the machine's left ventricle.
1215736865 M * Bertl darkpsi: if bind would be able to take down the host's devices, I would know :)
1215736925 M * darkpsi i'm not sure it's taking the hosts devices down, that's an uninformed hunch
1215736931 M * darkpsi i do know my machine is begging me for mercy
1215737085 M * darkpsi btw, the gentoo bind i'm emerging is 9.41_p1
1215737105 M * darkpsi defaults, +berkdb +ssl
1215737160 M * darkpsi if this does the same as on the x86_64, i'll lose my internet connection :)
1215737169 M * darkpsi and you'll be done with me ;)
1215737462 M * darkpsi bind bind bind bind...
1215737519 M * sebastiandeutsch i asked it two days ago - but again - i'm on etch and i dont want the vservers to go in /var/lib/vservers is --pkgbase my friend to choose a different location?
1215737544 M * darkpsi util-vserr seemd to obey the --prefix option for me
1215737625 M * daniel_hozac no, you want to use --rootdir.
1215737642 M * darkpsi sebastian: bad typing, sorry.  util-vserver doesn't have a pkgbase, but does do prefix correctly
1215737648 M * daniel_hozac i mean, --pkgbase is also necessary if you want to use external package management.
1215737654 J * _gh_ ~gerrit@c-67-169-199-103.hsd1.or.comcast.net
1215737671 M * darkpsi ./configure--help | grep pkg   comes up empty
1215737690 M * daniel_hozac it's not a ./configure option...
1215737700 M * darkpsi eh i see. missed the debian reference.
1215737706 M * darkpsi been a while since i touched debian
1215737713 M * sebastiandeutsch ;-)
1215737719 M * sebastiandeutsch what are u workin on?
1215737741 M * darkpsi gentoo, and, compiled from scratch
1215737757 M * darkpsi gentoo at home, compiled from scratch in production
1215737792 M * sebastiandeutsch maybe I consider gentoo too
1215737801 M * sebastiandeutsch currently I'm just playing around
1215737807 M * darkpsi gentoo is nice, not for playing around
1215737812 M * darkpsi stick with debian for playing around
1215737818 M * darkpsi for pretty colors, try suse
1215737887 M * darkpsi debian is by far the fastest off-the-ground, since everything's compiled already, and apt does a good job (most of the time)
1215737895 M * darkpsi bind finished..
1215737970 M * darkpsi ...
1215737979 A * darkpsi sighs
1215737990 A * darkpsi emerged bind into the vhive
1215738080 M * darkpsi okay. mystically, defaults are different. net-dns/bind-9.4.1_p1  USE="berkdb ipv6 ssl 
1215738299 Q * _gh_ Ping timeout: 480 seconds
1215738319 M * sebastiandeutsch i updated my questions in the wiki ;-) http://linux-vserver.org/util-vserver:Useful_commands
1215738569 Q * darkpsi Read error: Connection reset by peer
1215738810 J * darkpsi ~no@72-48-199-234.dyn.grandenetworks.net
1215738866 M * Bertl sebastiandeutsch: excellent! tx!
1215738991 N * Guest11 Genghis
1215739026 N * Genghis Guest16
1215739034 M * sebastiandeutsch can I also ask a vserver usage question here?
1215739104 M * daniel_hozac it's what we're here for...
1215739156 M * sebastiandeutsch (coming from php.pecl channel i'm surprised with this breeze of fresh air)
1215739173 M * sebastiandeutsch i created a vserver image, installed openssh server
1215739188 M * daniel_hozac http://linux-vserver.org/Frequently_Asked_Questions#When_I_try_to_ssh_to_the_guest.2C_I_log_into_the_host.2C_even_if_I_installed_sshd_on_the_guest._What.27s_wrong_here.3F
1215739231 A * cehteh somewhere posted an hint how to chbind the host server .. rather uncommon but handy
1215739242 M * sebastiandeutsch damn u're good
1215739250 M * sebastiandeutsch thanks again
1215739253 M * cehteh generally you just want to reconfigure the sshd
1215739263 Q * FireEgl Read error: No route to host
1215739494 J * FireEgl FireEgl@adsl-4-50-84.bhm.bellsouth.net
1215741159 J * _gh_ ~gerrit@67.170.155.50
1215741393 J * __gh__ ~gerrit@67.170.155.50
1215741807 Q * _gh_ Ping timeout: 480 seconds
1215742651 N * Guest16 Genghis
1215742686 N * Genghis Guest25
1215746156 P * sebastiandeutsch 
1215746311 N * Guest25 Genghis
1215746346 N * Genghis Guest36
1215749409 M * Bertl daniel_hozac: here is the promised patch for the flag changes: http://vserver.13thfloor.at/Experimental/delta-vflags-feat01.diff
1215749443 M * Bertl will test and integrate it into a new release tomorrow ... please let me know what you think and what patches you might want to see in the next experimental release
1215749479 M * Bertl (this fixes all currently supported filesystems, AFAICT)
1215749517 M * Bertl I'm off to bed now ... so have a good one everyone!
1215749522 N * Bertl Bertl_zZ
1215749971 N * Guest36 Genghis
1215750006 N * Genghis Guest41
1215751550 Q * nenolod Read error: Connection reset by peer
1215751683 J * nenolod ~nenolod@ip70-189-74-62.ok.ok.cox.net
1215753048 J * mOEnqomv_ ~hollow@proteus.croup.de
1215753298 J * micah_ ~micah@micah.riseup.net
1215753362 Q * Aiken resistance.oftc.net larich.oftc.net
1215753362 Q * fatgoose resistance.oftc.net larich.oftc.net
1215753362 Q * tzanger resistance.oftc.net larich.oftc.net
1215753362 Q * Hollow resistance.oftc.net larich.oftc.net
1215753362 Q * infowolfe resistance.oftc.net larich.oftc.net
1215753362 Q * micah resistance.oftc.net larich.oftc.net
1215753365 J * tzanger ~tzanger@gromit.mixdown.ca
1215753407 N * mOEnqomv_ Hollow
1215753560 J * Aiken ~james@ppp121-45-243-126.lns2.bne4.internode.on.net
1215753560 J * fatgoose ~samuel@82.80.modemcable.oricom.ca
1215753560 J * infowolfe ~infowolfe@c-67-160-149-42.hsd1.or.comcast.net
1215753566 Q * FloodServ resistance.oftc.net synthon.oftc.net
1215753571 N * Guest41 Genghis
1215754310 T * ChanServ http://linux-vserver.org/ |stable 2.2.0.7, devel 2.3.0.34, grsec 2.2.0.7|util-vserver-0.30.215|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 forget about the minute. 
1215754310 J * FloodServ services@services.oftc.net
1215754352 N * Genghis Guest52
1215755299 J * phedny ~mark@2001:610:656::115
1215756865 Q * darkpsi 
1215758016 N * Guest52 Genghis
1215758051 N * Genghis Guest55
1215758170 Q * besonen_mobile Read error: Connection reset by peer
1215758754 J * Slydder ~chuck@194.59.17.53
1215760481 J * dna ~dna@21-238-dsl.kielnet.net
1215761214 J * mire ~mire@8-168-222-85.adsl.verat.net
1215761227 Q * mire 
1215761263 J * pisco ~pisco@tor.noreply.org
1215761676 N * Guest55 Genghis
1215761711 N * Genghis Guest60
1215761811 N * DoberMann[ZZZzzz] DoberMann
1215762660 N * Guest60 Genghis
1215762830 J * kir ~kir@swsoft-msk-nat.sw.ru
1215763329 Q * Slydder Quit: Leaving.
1215763477 Q * balbir Ping timeout: 480 seconds
1215763512 J * Slydder ~chuck@194.59.17.53
1215763527 Q * Slydder 
1215764128 J * Slydder ~chuck@194.59.17.53
1215764538 Q * larsivi Quit: Konversation terminated!
1215765252 Q * eSa| Quit: KVIrc 3.4.0 Virgo http://www.kvirc.net/
1215766180 Q * harry Ping timeout: 480 seconds
1215766631 J * loddafnir ~mike@193.170.138.233
1215767963 J * harry ~harry@d51A461B4.access.telenet.be
1215770534 J * larsivi ~larsivi@85.221.53.194
1215771012 Q * arekm Read error: Connection reset by peer
1215771169 J * arekm arekm@carme.pld-linux.org
1215771511 Q * ktwilight Read error: Connection reset by peer
1215771538 J * ktwilight ~ktwilight@153.98-66-87.adsl-dyn.isp.belgacom.be
1215773376 Q * pisco Ping timeout: 480 seconds
1215774026 J * pisco_ ~pisco@tor.noreply.org
1215774812 J * yarihm ~yarihm@whitehead2.nine.ch
1215775325 Q * yarihm Ping timeout: 480 seconds
1215776144 J * yarihm ~yarihm@whitehead2.nine.ch
1215776394 J * sebastiandeutsch ~sebastian@i53876573.versanet.de
1215776607 J * dreamind ~dreamind@mx01.ap-wdsl.de
1215776624 M * dreamind Hi folks :)
1215776663 M * dreamind can anybody tell me wether I need to patch the kernel (or only vserver-utils) to get more than 16 ips per vserver?
1215777030 J * PeMPeK ~sign@118.137.201.253
1215777049 Q * PeMPeK 
1215777595 M * harry dreamind: you need to patch kernel and utils
1215777610 M * harry NB_IPV4ROOT or something like that
1215777618 M * dreamind ok
1215777630 M * dreamind in kernel and in utils that definition? not more?
1215778252 N * Bertl_zZ Bertl
1215778263 M * Bertl morning folks!
1215778272 Q * Slydder Quit: Leaving.
1215778291 M * Bertl dreamind: with recent (devel) patches, the limitation is gone
1215778958 M * daniel_hozac the utils haven't had that limitation either for quite some time.
1215778980 Q * larsivi Quit: Konversation terminated!
1215778997 M * Bertl hey daniel_hozac! how do you like my (last) patch?
1215779025 M * dreamind Bertl: ok
1215779028 M * daniel_hozac just reading it now.
1215779037 M * Bertl ah, excellent! np, take your time
1215779040 M * dreamind Bertl: but I use the stable version with 2.6.22.19
1215779090 M * daniel_hozac Bertl: what's the reason for the separate V_COW flag? (i guess that might become apparent soon... will keep reading)
1215779236 M * Bertl daniel_hozac: well, atm it isn't really used, it's more 'reserved'
1215779267 M * Bertl daniel_hozac: the idea is to make the cow link breaking independant from ixunlink and immutable files
1215779742 M * daniel_hozac Bertl: i'm just concerned about the new bit values...
1215779756 M * Bertl which ones?
1215779757 M * daniel_hozac is there a reason we can't keep the on-disk layout the same?
1215779763 M * daniel_hozac JFS, in this case.
1215779799 M * Bertl well, I decided to change the bit values to synchronize them between the filesystems (ext), and to move them upwards, to avoid future collisions
1215779836 M * daniel_hozac okay.... well, i guess it's acceptable, XFS has the same problem anyhow.
1215779839 M * Bertl for xfs we need a new layout anyways, so I thought, we make a big note for the upcoming release, that the filesystem attributes changed
1215779848 M * daniel_hozac right.
1215779882 M * Bertl IMHO the barrier is the only critical part, but if we can 'migrate' the cow stuff with a script, then even better
1215779971 M * daniel_hozac i was just thinking that. it shouldn't be hard to dump the flags and restore them...
1215781175 P * dreamind 
1215781482 N * ace_ sannes
1215781633 Q * Aiken Quit: Leaving
1215781650 Q * fatgoose Read error: Connection reset by peer
1215781720 J * fatgoose ~samuel@82.80.modemcable.oricom.ca
1215782255 M * daniel_hozac Bertl: i think the patch looks good.
1215782365 M * Bertl good, any pending patches on your side?
1215782374 M * daniel_hozac Bertl: is there a reason we use PDEs for the hide_check?
1215782399 M * Bertl IIRC, it was because of the dynamic inode allocation
1215782424 M * daniel_hozac so the PDE is persistent, the PROC_I is not?
1215782438 M * Bertl something like that, might have changed by now
1215782481 M * daniel_hozac i don't have anything merge ready yet.
1215782496 M * Bertl okay, no problem
1215782502 M * daniel_hozac (http://people.linux-vserver.org/~dhozac/p/k/delta-signal-fix04.diff is in the latest tree, right?)
1215782523 M * Bertl I'll double check, but I think yes
1215782677 M * Bertl nah, seems to be missing ...
1215783026 M * daniel_hozac hmm.
1215783117 M * daniel_hozac i don't understand vx_hide_check. the arguments it calls vx_check with seem... wrong.
1215783187 M * daniel_hozac or at least the second argument.
1215783202 M * Bertl it's always check/mask
1215783255 M * daniel_hozac but vx_check checks the context, and allows things based on the second argument.
1215783269 M * daniel_hozac i think m should be VS_ADMIN|VS_WATCH or so.
1215783311 M * daniel_hozac and not IATTR_* (whatever happens to be set)
1215783342 M * Bertl well, those use the same bit-numbers, but yes, valid point
1215783378 M * Bertl no problem if you break out a separate __ia_check()
1215783385 M * Bertl (with limited checks)
1215784210 Q * sannes Ping timeout: 480 seconds
1215784526 Q * jsambrook Quit: Leaving.
1215784561 J * jsambrook ~jsambrook@aelfric.plus.com
1215785436 M * daniel_hozac Bertl: http://people.linux-vserver.org/~dhozac/p/k/delta-procsec-feat05.diff
1215785464 M * daniel_hozac seems to work fine.
1215785478 M * daniel_hozac but i'll test some more with regard to persistence and such.
1215786277 M * Bertl +ei->vx_flags |= IATTR_WRITE; ?
1215786287 M * daniel_hozac for the per-process entries.
1215786305 M * daniel_hozac i believe this will fix the loginuid problem.
1215786374 M * Bertl okay, makes sense
1215788493 Q * yarihm Quit: This computer has gone to sleep
1215788916 Q * pisco_ Ping timeout: 480 seconds
1215789964 N * micah_ micah
1215790580 M * yang hello Bertl I am wondering if its possible to create a VPN connection on a guest or can it only be done through host ?
1215790684 J * sannes ace@har.sagt.no
1215790730 J * michal ~michal@www.rsbac.org
1215790733 M * michal hey :)
1215790857 J * raphinou ~rb@88.197.235.173
1215790868 M * michal daniel_hozac: ping
1215790896 M * daniel_hozac pong
1215790933 M * michal hey, i'm installing centos+vserver, tell me please, is your repo current?
1215790965 M * Bertl yang: you can _use_ any vpn connection inside a guest
1215790968 M * michal just added description to yum.repos.d but yum search does not show any kind of package from there
1215790982 M * Bertl yang: by default, you cannot _create_ one, (similar to new interfaces)
1215791036 M * daniel_hozac michal: what kernel do you have installed?
1215791048 M * raphinou hi. What's the best way to share a small number of files bewteen multiple vservers. I'm thinking eg of /etc/hosts, which would have one line for each vserver, and each vserver needing the same entries. Might unify be to heavy for this?
1215791053 M * michal current one from centos, YET. 
1215791057 M * michal 5.2 btw
1215791065 M * daniel_hozac that's still 2.6.18, right?
1215791069 M * michal yes
1215791081 M * michal i'm going to install one from repo - if yum is going to see it ;]
1215791093 M * daniel_hozac and you're sure you added the repo, and it's enabled?
1215791100 N * DoberMann DoberMann[PullA]
1215791110 M * michal yes
1215791117 M * daniel_hozac so yum repolist shows it?
1215791141 M * michal dhzac-vserver, description, enabled
1215791147 M * michal looks ok :>
1215791165 M * daniel_hozac do you have protectbase installed?
1215791183 M * michal no idea, checking on
1215791202 M * michal yes, i gues that's the one guilty here
1215791205 M * daniel_hozac yep.
1215791414 M * fosco hi
1215791416 M * yang Bertl: I never used openvpn before, but to start it it requires some OpenVPN requires TUN/TAP driver support in the kernel. You'll also need a
1215791421 M * yang tun device file. 
1215791421 M * fosco I have a problem when starting a vserver
1215791427 M * fosco on a new install
1215791443 M * fosco # vserver toto start
1215791447 M * fosco vsched: readdir(): Bad address
1215791456 M * yang Bertl: the thing is I would like to connect from a remote server into vserver-guest will this work, or shoudl it be done only on host?
1215791456 M * daniel_hozac fosco: what architecture?
1215791460 M * fosco sparc64
1215791463 M * Bertl yang: yep, both can be pre-created on the host (persistant tun) and used by the guest
1215791471 M * yang nice
1215791492 M * yang Bertl: any manuals for this on the vserver website?
1215791492 M * daniel_hozac fosco: http://people.linux-vserver.org/~dhozac/p/m/delta-dietdirent-fix01.diff
1215791500 M * fosco thank you :)
1215791514 M * daniel_hozac apply to dietlibc, rebuild it, rebuild util-vserver with the new diet, and you should be golden.
1215791568 M * fosco ok...
1215791653 M * yang Bertl: found it !
1215791709 M * fosco daniel_hozac: is util-vserver statically linker to dietlibc?
1215791725 M * daniel_hozac yes.
1215791731 M * daniel_hozac that's kind of the idea behind dietlibc :)
1215791733 M * Bertl yang: IIRC, daniel_hozac is/was working on some automatic tun interface setup
1215791735 M * fosco 'k
1215791743 M * daniel_hozac Bertl: it's done. 0.30.215 has everything.
1215791751 M * fosco so I need to rebuild the packages of dietlibc and then util-vserver
1215791752 M * Bertl yang: hear?
1215791764 M * daniel_hozac fosco: right.
1215791775 M * yang http://linux-vserver.org/Frequently_Asked_Questions#Can_I_run_an_OpenVPN_Server_in_a_guest.3F
1215791818 M * Bertl yang: so you can replace the 'host' magic by vserver config now
1215791942 M * yang Bertl: so, should I follow the manual or is it done differently now?
1215791973 M * daniel_hozac http://paste.linux-vserver.org/12311 is how my VPN-server is setup.
1215791996 J * Mojo1978 ~Mojo1978@ip-78-94-103-112.hsi.ish.de
1215792027 M * Bertl yang: best would be to try the 'new' way, and add it to the wiki (once you figured the details)
1215792088 Q * raphinou Quit: Leaving
1215792731 M * fosco daniel_hozac: thank you
1215792757 M * fosco that's pretty easy to fix, but I could have searched for a very loooong time :)
1215792801 Q * mick_home Max SendQ exceeded
1215792832 M * daniel_hozac yeah, it wasn't all that obvious when we were tracking it down.
1215792839 M * fosco I imagine
1215792850 M * fosco that's pretty complex to find :)
1215793024 P * kir Leaving.
1215794420 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d
1215794700 J * snobbo ~Shane@125.163.224.248
1215794708 P * snobbo Noneya bussiness!
1215795460 J * yarihm ~yarihm@84-74-147-84.dclient.hispeed.ch
1215796034 J * marv_ ~marv@modemcable174.71-82-70.mc.videotron.ca
1215797533 J * awake ~r5ck4wef@72-160-216-154.dyn.centurytel.net
1215797853 Q * awake Quit: awake has no reason
1215797920 N * DoberMann[PullA] DoberMann
1215798735 J * awake ~irc-user@72-160-216-154.dyn.centurytel.net
1215799400 Q * sebastiandeutsch Quit: sebastiandeutsch
1215799443 J * sebastiandeutsch ~sebastian@i53876573.versanet.de
1215800186 P * sebastiandeutsch 
1215802775 Q * phedny Ping timeout: 480 seconds
1215803085 J * phedny ~mark@2001:610:656::115
1215803392 Q * dna Ping timeout: 480 seconds
1215804754 J * sebastiandeutsch_ ~sebastian@i538767EF.versanet.de
1215804853 J * dna ~dna@21-238-dsl.kielnet.net
1215805600 P * sebastiandeutsch_ 
1215806203 Q * awake Ping timeout: 480 seconds
1215807473 J * sebastiandeutsch ~sebastian@i538767EF.versanet.de
1215807561 P * sebastiandeutsch 
1215808040 N * DoberMann DoberMann[ZZZzzz]
1215808121 Q * loddafnir Ping timeout: 480 seconds
1215809291 M * Bertl okay, off for now .. bbl
1215809297 N * Bertl Bertl_oO
1215811247 J * larsivi ~larsivi@169.80-202-217.nextgentel.com
1215811667 Q * dna Ping timeout: 480 seconds
1215812932 J * awake ~irc-user@72-160-217-79.dyn.centurytel.net
1215813215 Q * yarihm Ping timeout: 480 seconds
1215813810 J * the_fafa ~fafa@p5496F23B.dip.t-dialin.net
1215814221 Q * fafa_ Ping timeout: 480 seconds
1215815205 J * yarihm ~yarihm@84-74-147-84.dclient.hispeed.ch
1215816086 J * Aiken ~james@ppp121-45-243-126.lns2.bne4.internode.on.net
1215817690 Q * bonbons Quit: Leaving
1215817715 Q * yarihm Ping timeout: 480 seconds
1215819899 Q * awake Ping timeout: 480 seconds
1215820034 J * awake ~irc-user@75-121-135-186.dyn.centurytel.net