1174694896 Q * dghill Quit: Weeeeee!
1174694899 J * dghill dghill@office.mel.illuminate.com.au
1174694915 Q * dghill 
1174694921 J * dghill dghill@office.mel.illuminate.com.au
1174695241 N * DoberMann DoberMann[ZZZzzz]
1174698219 Q * jkl Quit: rebuilding raid array
1174698718 J * jkl maGAILnb2b@c-67-173-253-237.hsd1.co.comcast.net
1174700437 Q * yarihm Quit: Leaving
1174700705 M * Bertl doener: still around?
1174701558 J * lylix ~eric@dynamic-acs-24-154-33-9.zoominternet.net
1174701687 M * Bertl wb lylix!
1174701748 M * lylix Bertl: g/e... 
1174701767 M * lylix well, updates on kernel did wonders... tnx fellas
1174701776 M * Bertl everything fine now with udp, I take :)
1174701781 M * lylix all cool
1174701793 M * Bertl excellent, we were worried with you for a moment :)
1174701811 M * lylix turned out to become a major problem as of late
1174701819 M * lylix im surprised it was not catch sooner
1174701827 M * lylix s/catch/caught
1174701948 M * Bertl some things stay dormant for quite a while, then suddenly they pop up ...
1174703281 Q * shedi Ping timeout: 480 seconds
1174703880 J * shedi ~siggi@ftth-237-144.hive.is
1174705116 Q * softi42 Ping timeout: 480 seconds
1174705604 M * ntrs Hi everyone.
1174705610 M * Bertl hey ntrs!
1174705684 M * ntrs Someone wants to install some software inside a guest and the software requires the following:
1174705686 M * ntrs kernel.shmax = 268435456
1174705686 M * ntrs kernel.msgmni = 1024
1174705686 M * ntrs kernel.sem = 1024
1174705704 M * ntrs Is there a way to provide that or fake it inside a guest?
1174705715 J * softi42 ~softi@p549d7088.dip.t-dialin.net
1174705720 M * Bertl depends on the kernel version
1174705781 M * ntrs that host runs 2.6.16.16 with 2.0.2-rc20
1174705803 M * ntrs how does it depend?
1174705819 M * Bertl well, recent kernels (2.6.19/20) have ipc namespace support
1174705834 M * Bertl you can configure those values there via the sysctl config
1174705847 M * ntrs on the host or the guest?
1174705857 M * Bertl in the guest config _on_ the host
1174705884 M * ntrs what if I make the change on the host only? It will not be visible in the guest?
1174705907 M * Bertl for the older kernels, the host settings will show up in the guest too
1174705926 M * Bertl but that doesn't mean that the requested amount is actually available
1174705946 M * ntrs I think I can set the semaphores setting in the guest config even with this kernel version, right?
1174705962 M * ntrs which setting is the one that will not work?
1174705974 M * Bertl the limits, yes, but not the kernel side interpretation
1174706002 M * Bertl i.e. it is handled differently
1174706078 M * ntrs so, I guess I could just set it on the host now
1174706100 M * ntrs what is .shmax?
1174706105 M * Bertl yep, and if you set it a little larger, it might actually provide that much data
1174706115 M * Bertl maximum of shared memory
1174706124 M * Bertl means 256MB
1174706143 M * ntrs why would any software need that value set?
1174706153 M * ntrs when if unset, it can use all of it.
1174706162 M * Bertl well, if it locks down 256MB of your memory?
1174706219 M * Bertl I have no idea what software except for large databases and number chrunching would need that though
1174706226 M * ntrs do the above figures look safe?
1174706236 M * ntrs yes, this is needed by IBM's DB2
1174706248 M * Bertl how much memory do you 'assign' to that guest?
1174706271 M * ntrs 512 MB rss
1174706287 M * Bertl okay, then it is reasonable that it can lock/share 256MB
1174706295 M * ntrs ok
1174706310 M * ntrs the msgmni=1024, is that safe?
1174706359 M * Bertl maximum number of message queues
1174706374 M * Bertl a little high, but shouldn't hurt ...
1174706383 M * Bertl note: default is 50
1174706408 M * ntrs ok
1174706458 M * Bertl hmm, no, sec
1174706481 M * Bertl default is 16
1174706515 M * ntrs I'll give it a try and will see what happens.
1174707560 Q * FireEgl Quit: ...
1174709204 Q * ensc Killed (NickServ (GHOST command used by ensc_))
1174709213 J * ensc ~irc-ensc@p54b4f337.dip.t-dialin.net
1174710434 J * FireEgl ~FireEgl@adsl-61-136-122.bhm.bellsouth.net
1174712989 J * Aiken ~james@ppp250-73.lns2.bne4.internode.on.net
1174713536 J * DoberMann_ ~james@AToulouse-156-1-21-14.w86-196.abo.wanadoo.fr
1174713646 Q * DoberMann[ZZZzzz] Ping timeout: 480 seconds
1174713850 Q * wenchien Quit: Terminated with extreme prejudice - dircproxy 1.0.5
1174715605 J * wenchien ~wenchien@59-105-176-11.adsl.static.seed.net.tw
1174718094 M * Bertl okay, off to bed ... have a good one everyone! cya!
1174718100 N * Bertl Bertl_zZ
1174720065 Q * arachnist Quit: bbl
1174721280 Q * Aiken Quit: Leaving
1174721307 J * boci^ boci@pool-4774.adsl.interware.hu
1174723711 Q * wenchien Quit: Terminated with extreme prejudice - dircproxy 1.0.5
1174724524 J * wenchien ~wenchien@59-105-176-11.adsl.static.seed.net.tw
1174724858 J * dna ~naucki@p54bcdb6c.dip.t-dialin.net
1174725662 N * DoberMann_ DoberMann
1174725821 J * arachnist ~arachnist@088156185052.who.vectranet.pl
1174726415 Q * arachnist Quit: Reconnecting
1174726417 J * arachnist ~arachnist@088156185052.who.vectranet.pl
1174726740 J * Aiken ~james@ppp250-73.lns2.bne4.internode.on.net
1174728296 M * arachnist hmm
1174728409 M * arachnist how does one set /etc/vservers/newnest/vprocunhide-files ? one file per line?
1174728542 Q * FireEgl Quit: ...
1174728622 J * FireEgl Proteus@adsl-61-136-122.bhm.bellsouth.net
1174728679 M * daniel_hozac yes, but it's /etc/vservers/.defaults/apps/vprocunhide/files
1174728689 M * daniel_hozac (as per the flower page)
1174728775 M * arachnist ah
1174729121 M * arachnist strange, i have '/proc/bus/' in /etc/vservers/.defaults/apps/vprocunhide/files and i can access /proc/bus/pci in vserver, but can't /proc/bus/usb
1174729141 M * daniel_hozac because /proc/bus/usb is a separate filesystem.
1174729146 M * arachnist mhm
1174729153 M * arachnist makes sense
1174730661 M * arachnist hmm, i don't get it. i've put /proc/bus/usb/ in the /etc/vservers/.defaults/apps/vprocunhide/files, ran /usr/lib/util-vserver/vprocunhide, but then i got a lot of "Invalid argument" (one for every file in /proc/bus/usb)
1174730717 M * arachnist hmm
1174731314 M * daniel_hozac yep.
1174731327 M * daniel_hozac since /proc/bus/usb is a separate filesystem, it doesn't support those flags.
1174731393 J * bonbons ~bonbons@83.222.39.201
1174731642 N * DoberMann DoberMann[PullA]
1174731683 M * arachnist is there a patch for it or something?
1174731837 M * daniel_hozac for what?
1174731844 M * daniel_hozac unhiding /proc/bus/usb?
1174731846 M * arachnist yup
1174731950 M * daniel_hozac http://archives.linux-vserver.org/200610/0054.html
1174732023 M * sannes hm, IPv6 patch doesn't apply to rc19 because of the addition of if (!dev) goto out;, guess that should be kept :) (<-- for whoever maintains ipv6 support patch for vs2.2.0)
1174732038 M * daniel_hozac yep, i know.
1174732058 M * daniel_hozac bonbons: you want to update it, or should i?
1174732171 M * arachnist daniel_hozac: thanks
1174732191 M * bonbons daniel_hozac: I don't mind who updates first :)
1174732415 M * sannes :)
1174732453 M * daniel_hozac well, i guess it's better if you do it as you have that fancy page ;)
1174732476 M * bonbons ok, will do
1174733807 M * bonbons daniel_hozac: do you know if Bertl plans rc20 for 2.6.20.4? (just two files with minor offset, ignoring the standard reject in Makefile)
1174733873 M * daniel_hozac i don't think the issues with rc19 are resolved yet.
1174733901 M * bonbons ok
1174734313 M * arachnist now that's strange. i can ssh into vserver, but can't open xterm/konsole while in
1174734333 M * arachnist thought that when one works, the other should also
1174734349 M * daniel_hozac is the X forwarding working as it should?
1174734372 M * daniel_hozac e.g. the cookie set correctly, DISPLAY, etc.
1174734374 M * arachnist i don't use X forwarding, i'm running X server inside vserver
1174734406 M * daniel_hozac uh, so you ssh into the guest to run things on the X server which is also running on it?
1174734464 M * arachnist no, i ssh into the vserver to do some stuff, and do a 'vserver newnest exec su - arachnist -c "/usr/bin/startx"' as root directly on host
1174734577 M * arachnist hmm
1174734586 M * daniel_hozac so, is DISPLAY set correctly?
1174734589 M * arachnist yes
1174734610 M * daniel_hozac and vserver newnest exec su - arachnist -c "xterm" works fine?
1174734616 M * arachnist X works ok, i just can't open any *term in it, and i think i've found the cause
1174734632 M * arachnist i should probably add some pty devices to /dev in vserver ;>
1174734640 M * daniel_hozac bad permissions on /dev/ptmx or /dev/pts?
1174734648 M * daniel_hozac /dev/pty* is legacy...
1174734659 M * arachnist crw-r--r-- 1 root root   5,   2 2007-03-24 12:11 ptmx
1174734675 M * arachnist should be 666?
1174734680 M * daniel_hozac usually, yes.
1174734687 M * arachnist ok, will try
1174734771 M * arachnist yay, it works :)
1174734777 M * arachnist thanks :>
1174734819 M * sannes What issues is there with vs2.2.0-rc19?
1174734910 M * daniel_hozac oops in do_task_stat, vps possibly showing the wrong init, and one more i can't think of right now...
1174734929 M * sannes And earlier versions do not have this problem?
1174734967 M * sannes (oops is a showstopper I guess..)
1174735016 M * daniel_hozac the oops is most likely present in all 2.6.19/2.6.20 versions.
1174735049 M * daniel_hozac it's really hard to trigger though, as we haven't been able to reproduce it at will yet, and only seen it a few times.
1174735339 M * sannes so if I have been running rc16/17(+ipv6) with 2.6.20.3 I really don't have much to loose running rc19(+ipv6) with 2.6.20.4 .. ?
1174735458 M * daniel_hozac nope, you just gain the fixes made since then.
1174735589 M * sannes well, I guess that clears it up for me, just got a bit iffy when I saw "issues" :P
1174735986 J * b0c1 ~boci@pool-6880.adsl.interware.hu
1174736099 J * b0c1k4 ~boci@pool-6880.adsl.interware.hu
1174736300 Q * boci^ Ping timeout: 480 seconds
1174736307 J * boci^ ~boci@pool-2154.adsl.interware.hu
1174736539 Q * b0c1 Ping timeout: 480 seconds
1174736600 Q * b0c1k4 Ping timeout: 480 seconds
1174736894 Q * bonbons Remote host closed the connection
1174736987 J * b0c1 ~boci@pool-2412.adsl.interware.hu
1174737143 J * b0c1k4 ~boci@pool-2412.adsl.interware.hu
1174737235 J * bonbons ~bonbons@83.222.39.201
1174737416 Q * boci^ Ping timeout: 480 seconds
1174737445 J * boci^ ~boci@pool-2412.adsl.interware.hu
1174737445 Q * boci^ Read error: Connection reset by peer
1174737483 J * boci^ ~boci@pool-6431.adsl.interware.hu
1174737594 Q * b0c1 Ping timeout: 480 seconds
1174737714 Q * b0c1k4 Ping timeout: 480 seconds
1174737762 Q * boci^ 
1174738361 Q * Aiken Quit: Leaving
1174738656 J * boci^ ~boci@pool-6431.adsl.interware.hu
1174738690 N * DoberMann[PullA] DoberMann
1174740106 J * shenjian ~shenjian@222.93.212.250
1174740417 P * shenjian Leaving
1174740697 Q * FireEgl Remote host closed the connection
1174740837 M * bonbons uploaded updated patch ( http://people.linux-vserver.org/~bonbons/ipv6/patch-2.6.20.4-vs2.2.0.ipv6-rc19.diff )
1174740892 M * bonbons note: when patching 2.6.20.4 with 2.6.20.3-vs2.2.0-rc19 there are a few offsets and usual reject for Makefile
1174740893 M * daniel_hozac is there any reason you don't just the .ipv6 at the end, rather than in the middle?
1174740927 M * bonbons not really
1174741253 M * daniel_hozac bonbons: btw, do you still have the host that was having problems with vshelper?
1174741438 M * bonbons yes, there were a few of them, but only one of them here at home
1174741524 M * daniel_hozac http://people.linux-vserver.org/~dhozac/p/uv/experimental/delta-vshelper-reboot.diff should fix reboot in a (IMHO) safe way too.
1174743182 M * bonbons first upgraded to rc5, there depending on the init style I get two scenarios: standard gentoo init framework works fine, inittab only does not work (e.g. 30 sec timeout with 100% sys CPU load)
1174743190 M * bonbons now checking with applied patch...
1174743284 M * daniel_hozac only on reboot though, right?
1174743290 M * daniel_hozac i.e. halt is fine
1174743432 M * bonbons hmm, did just check reboot with the trimmed-down init (halt to be checked again later - have already patched now)
1174743474 M * bonbons the full gentoo init system was working well (always with real init inside guest), halt or reboot
1174743580 M * bonbons hm, with patched version it waits for something on vserver <guest> start after doing shutdown -h now from inside... (and guest being stopped)
1174743630 M * bonbons cause: some vshelper sleeping on it's lockfile
1174743649 M * bonbons   /bin/cat /tmp/vshelper-stop-sync.aqNvC6/pipe
1174743844 M * bonbons reboot is fine for that one, same issue for the normal guest
1174743922 M * daniel_hozac hmm, odd...
1174744142 M * bonbons will continue testing after lunch
1174744993 M * arachnist strange thing, in vserver i get ~16k fps with glxgears, outside ~18k fps. but going back is not worth those 2k fps
1174745081 M * daniel_hozac chroot /vservers/guest glxgears also gives you 18k?
1174745087 M * arachnist yup
1174745130 M * daniel_hozac and you do it as the same user both times?
1174745166 M * arachnist well, different user account, but shared ~/ and  and identical groups
1174745285 N * Bertl_zZ Bertl
1174745295 M * Bertl http://wiki.cchtml.com/index.php/Glxgears_is_not_a_Benchmark
1174745340 M * Bertl morning folks!
1174745354 M * daniel_hozac morning Bertl!
1174745660 P * lylix 
1174745712 M * Bertl daniel_hozac, doener: I have a new theory for the tty issue
1174745716 M * daniel_hozac Bertl: oh?
1174745761 M * Bertl what made me think is that we a) couldn't really recreate it, b) it wasn't seen on mainline (yet) and c) we have that issue since 2.6.19 (as it seems)
1174745818 M * daniel_hozac okay..
1174745831 M * Bertl so I gave it a big thought what the common denominator is, and I think it might be related with the pid=1 issue we have been postponing ...
1174745854 M * Bertl the idea goes like this:
1174745880 M * Bertl - in mainline, we explicitely purge a proc entry when a task goes away
1174745914 M * Bertl - assumed a guest init goes away, that happens too, but only for the real pid
1174745941 M * Bertl - something running in xid=1 might still reach that task via pid=1
1174745954 M * Bertl  (at least for a short time)
1174746113 M * daniel_hozac ah, hmm, yeah, that sounds plausible.
1174746146 M * Bertl so I fixed the init issue for now with an evil hack :)
1174746213 M * daniel_hozac okay, what's that?
1174746253 M * Bertl sec, uploading shortly
1174746584 M * Bertl http://vserver.13thfloor.at/Experimental/delta-init-fix02.diff
1174746751 M * Bertl while I'm not happy with the way it works, I think it might be good to get it tested with this in, as it should basically eliminate the beforementioned conditions for the tty issue
1174746799 M * Bertl (read: if we get the tty oops with this patch, it can't be as easy as the init pid issue :)
1174746861 M * daniel_hozac hehe.
1174746881 M * Bertl btw, it is based on your idea as you can see
1174746964 M * daniel_hozac yeah.
1174747130 M * Bertl so what's your options, quick rc20 and field testing?
1174747232 M * daniel_hozac well, we don't have a way to trigger the issue yet, do we?
1174747281 M * Bertl nope
1174747322 M * daniel_hozac so an rc20 for field testing makes sense to me.
1174747338 M * Bertl okay, I still haven't tested the patch for the fake init case
1174747345 M * Bertl (i.e. blend through init)
1174747368 M * Bertl I'm not sure it works there as expected ... but it shouldn't be hard to adjust
1174747386 M * daniel_hozac should work fine, IMHO
1174747410 M * Bertl okay, could you give it a spin on your side too before we do the rc20?
1174747508 M * daniel_hozac already compiling ;)
1174747523 M * Bertl excellent! you're the man!
1174747545 M * Bertl btw, what are your plans for april?
1174747575 M * daniel_hozac school as usual, i suppose. why?
1174747591 M * Bertl so you will be around on a regular basis, yes?
1174747644 M * daniel_hozac yeah, should be.
1174747671 M * Bertl okay, because I'm on vacation in april and will only show up sporadiccally ...
1174747684 M * daniel_hozac ah, nice. where are you going?
1174747718 M * Bertl we are going to the US, doing some sight seeing all over the place
1174747728 M * daniel_hozac oh, cool.
1174747850 M * Bertl hmm, shouldn't chcontext --flag info_init work?
1174747860 M * daniel_hozac it should, at least with recent utils.
1174747894 M * Bertl maybe that's the problem, sec
1174748281 M * Bertl as I thought, listing an init guest makes the blend through init disappear in the next listing :(
1174748303 M * Bertl it pops up if you repeat the lookup though
1174748587 J * b0c1 ~boci@pool-6431.adsl.interware.hu
1174748700 M * Bertl daniel_hozac: hmm, in what cases would vcontext --xid 101 --migrate -- .. return with EINVAL?
1174748831 Q * mire Remote host closed the connection
1174749024 M * Bertl anyway ... translocating now ... will be back in two hours ...
1174749033 N * Bertl Bertl_oO
1174749050 Q * boci^ Ping timeout: 480 seconds
1174751472 J * b0c1k4 ~boci@pool-3747.adsl.interware.hu
1174751488 Q * b0c1k4 
1174751491 J * boci^ ~boci@pool-3747.adsl.interware.hu
1174751906 Q * b0c1 Ping timeout: 480 seconds
1174752094 Q * boci^ Ping timeout: 480 seconds
1174752859 J * manolisaxxx ~manolisa@81.202.53.101.dyn.user.ono.com
1174752901 P * manolisaxxx 
1174753271 M * daniel_hozac Bertl_oO: that's a really good question... didn't you get a function name?
1174753538 J * boci^ ~boci@pool-8151.adsl.interware.hu
1174754053 J * DoberMann_ ~james@AToulouse-156-1-16-188.w86-196.abo.wanadoo.fr
1174754159 Q * DoberMann Ping timeout: 480 seconds
1174754539 Q * duckx Remote host closed the connection
1174754705 J * duckx ~Duck@tox.dyndns.org
1174754964 Q * DoberMann_ Ping timeout: 480 seconds
1174755161 J * DoberMann ~james@AToulouse-156-1-8-3.w81-250.abo.wanadoo.fr
1174756197 Q * FloodServ charon.oftc.net services.oftc.net
1174756410 J * FloodServ services@services.oftc.net
1174758279 Q * DoberMann Ping timeout: 480 seconds
1174758794 J * DoberMann ~james@AToulouse-156-1-162-208.w90-38.abo.wanadoo.fr
1174759032 N * Bertl_oO Bertl
1174759041 M * Bertl back now ...
1174759070 M * Bertl daniel_hozac: I will check shortly ...
1174759954 Q * DoberMann Ping timeout: 480 seconds
1174760747 J * DoberMann ~james@AToulouse-156-1-119-114.w90-30.abo.wanadoo.fr
1174761319 Q * infowolfe Ping timeout: 480 seconds
1174761913 M * Bertl daniel_hozac: $ vcontext --xid 101 --migrate -- ps auxwww
1174761913 M * Bertl vcontext: vc_ctx_migrate(): Invalid argument
1174761931 M * Bertl vcmd -i 101 -BC ctx_migrate -- ps auxwww
1174761936 M * Bertl (works as expected)
1174761990 M * Bertl happens here with 0.30.213-rc4
1174762114 M * daniel_hozac hmm, okay. anything special there/about the context?
1174762151 M * daniel_hozac works fine with 0.30.213-rc5 here.
1174762219 Q * DoberMann Ping timeout: 480 seconds
1174762231 M * Bertl sec, I uplaod the entire sequence
1174762245 M * daniel_hozac what kernel, btw?
1174762287 J * infowolfe ~infowolfe@c-67-164-195-129.hsd1.ut.comcast.net
1174762358 M * Bertl http://paste.linux-vserver.org/1349
1174762395 M * Bertl 2.6.19.7-vs2.2.0-rc19.2
1174762431 M * daniel_hozac interesting..
1174762491 M * daniel_hozac okay, reproducible.
1174762499 M * Bertl *phew* :)
1174762517 M * Bertl I thought I was seeing ghosts here :)
1174762540 M * daniel_hozac aah.
1174762549 M * daniel_hozac yeah, the migrate tries to enter the namespaces too.
1174762573 M * daniel_hozac (to keep vc_ctx_migrate consitent across kernels)
1174762626 M * daniel_hozac (uts and IPC spaces, that is)
1174762673 M * Bertl hmm, okay, why does it say the migrate failed?
1174762688 M * daniel_hozac because to the util, the migrate is what's failing ;)
1174762706 M * Bertl aha, hmm ...
1174762720 M * daniel_hozac i.e. vcontext runs vc_ctx_migrate, which will enter the spaces on a spaces-capable kernel.
1174762721 M * Bertl why does it try to enter non assigned namespaces?
1174762740 M * daniel_hozac because vc_ctx_migrate used to enter the uts and IPC spaces before they existed ;)
1174762775 M * Bertl okay, is that considered a feature or a bug? *G* 
1174763026 M * daniel_hozac feature, IMHO.
1174763046 M * daniel_hozac as vc_ctx_create will also create and set those spaces.
1174763155 Q * infowolfe Ping timeout: 480 seconds
1174763373 M * Bertl daniel_hozac: hmm, okay ... wouldn't it be a much better feature if it a) would check what spaces are assigned (we might need a syscall command for that?) and b) you could select what spaces you want to be created/migrated?
1174763448 M * Bertl on a different note: what is the problem Stéphane is currently struggling with?
1174763454 M * daniel_hozac well, sure.
1174763503 M * daniel_hozac i'm not really sure.
1174763605 M * daniel_hozac the spaces are going to need some rather large changes to the APIs in util-vserver.
1174763626 M * Bertl np, I'm fine with the current implementation
1174763646 M * daniel_hozac for now i'd rather just keep it the same across the kernels.
1174763660 M * Bertl but we should keep the spaces in mind, as they will take over the existing isolations
1174763678 M * Bertl and we should make sure that the modularity is preserved
1174763701 M * daniel_hozac absolutely.
1174763744 M * Bertl so maybe move the space stuff in a separate vspace tool and call that (maybe even from vcontext) or so (sooner or later)
1174763783 M * daniel_hozac yeah.
1174764872 M * ag- let's say i have only eth0 with a public IP and lo with loopback, i'd like to setup vservers on 192.168.0.0/24, what are my options? i must add that the ethernet VLAN on eth0 is hostile
1174764911 M * ag- i thought there was a solution to this, but i can no longer find/remember where...
1174764968 M * daniel_hozac dummy0, or just use eth0 with iptables to block access to the guests.
1174765041 M * ag- hmm... dummy0, i think that was it...
1174765118 M * ag- thanks daniel_hozac :)
1174765665 Q * trippeh Ping timeout: 480 seconds
1174766037 J * infowolfe ~infowolfe@c-67-164-195-129.hsd1.ut.comcast.net
1174766449 Q * slack101 Ping timeout: 480 seconds
1174766628 J * DoberMann ~james@AToulouse-156-1-184-61.w90-38.abo.wanadoo.fr
1174766895 J * trippeh atomt@uff.ugh.no
1174767399 J * yarihm ~yarihm@84-74-16-109.dclient.hispeed.ch
1174767530 J * DreamerC_ ~dreamerc@125-225-97-143.dynamic.hinet.net
1174767939 Q * DreamerC Ping timeout: 480 seconds
1174768208 J * Aiken ~james@ppp250-73.lns2.bne4.internode.on.net
1174769197 J * frz ~frz@86.59.25.121
1174769229 M * Bertl welcome frz! morning Aiken!
1174769246 M * frz hello :)
1174769298 M * Aiken hi
1174769594 P * frz 
1174773243 M * doener Bertl: I'm back now
1174773295 M * doener so the tty issue resolved itself to be a tty-driver issue to be a initpid/proc bug?
1174773322 M * daniel_hozac maybe.
1174773393 M * Bertl doener: no idea ...
1174773406 M * Bertl doener: but it is currently my best theory
1174773427 M * Bertl unfortunately the initpid stuff still doesn't work as expected
1174773778 J * bronson ~bronson@adsl-76-202-197-80.dsl.pltn13.sbcglobal.net
1174775422 J * DavidS ~david@chello062178045213.16.11.tuwien.teleweb.at
1174775586 Q * pagano Ping timeout: 480 seconds
1174777449 M * matti Hi Bertl ;)
1174778311 M * Bertl hey matti!
1174778416 M * Bertl daniel_hozac, doener: ping?
1174778420 M * daniel_hozac pong
1174778446 M * Bertl ah, good, it seems I have a working version now with much less changes than before (to mainline code)
1174778460 M * daniel_hozac oh?
1174778462 M * Bertl I'd appreciate a test run and a validation
1174778472 M * daniel_hozac sure.
1174778486 M * Bertl at least it passes all my test here ...
1174778498 M * daniel_hozac sounds good then!
1174778511 M * Bertl yeah, almost too good to be true :)
1174778528 M * daniel_hozac let's hope it isn't ;)
1174778594 Q * bonbons Quit: Leaving
1174778770 M * Bertl http://vserver.13thfloor.at/Experimental/delta-init-fix02b.diff
1174778776 M * Bertl instead of fix02.diff
1174778796 Q * yarihm Ping timeout: 480 seconds
1174778800 M * Bertl (not sure we actually want to disable the revalidate though)
1174778844 M * daniel_hozac hmm, and this works?
1174778858 M * Bertl well, it seems so ...
1174778876 M * daniel_hozac hmm, interesting..
1174778949 M * Bertl http://paste.linux-vserver.org/1351
1174778956 M * Bertl (forget the first two lines)
1174778998 M * Bertl I was testing all sequences of the last six vcmd commands and the vps
1174779060 Q * boci^ Quit: Távozom
1174779365 J * yarihm ~yarihm@84-74-16-109.dclient.hispeed.ch
1174779631 Q * dna Quit: Verlassend
1174779668 M * daniel_hozac it does appear to work fine.
1174779725 M * Bertl glad that you can confirm that :)
1174779761 M * daniel_hozac a much nicer solution than the first one ;)
1174779773 M * Bertl yes, definitely
1174779783 M * Bertl but we should double check security 
1174779902 M * daniel_hozac seems fine too, i.e. i can't look at tasks through /proc which aren't in the right context.
1174779923 M * Bertl perfect .. so we have an rc20 then?
1174779942 M * daniel_hozac wasn't there a third issue?
1174779972 M * Bertl hmm ...any hints?
1174780099 M * daniel_hozac hmm, no... since i can't remember it at all, it can't have been too important.
1174780114 M * Bertl the lylix issue maybe?
1174780127 M * Bertl resolved itself with the upgrade :)
1174780248 J * FireEgl Proteus@adsl-61-136-122.bhm.bellsouth.net
1174780254 M * daniel_hozac ah, i believe it was the unhash problem.
1174780284 M * Bertl but that is fixed too, no?
1174780298 M * daniel_hozac but that should be fixed since rc19, so assuming delta-init-fix02b fixes the tty->driver issue, we should be good.
1174780311 M * Bertl okay, rc20 it is then ...
1174780365 M * Bertl and a version update to 2.6.20.4 IIRC
1174780407 M * daniel_hozac yeah.