1156896074 M * Bertl basically ext2/3 is wroking already, but they will benefit from that change too
1156896812 M * nayco Ok, I will take time to try. That will be better for my article ;-)
1156896817 M * nayco thanks for all
1156896825 M * Bertl you're welcome!
1156896835 M * nayco *nap attack*
1156896838 M * nayco good night !
1156896845 M * Bertl have a good one!
1156896866 Q * nayco Quit: ZzZzZzzzzzZzZzzzzZZzzzzZ
1156899044 Q * micah Server closed connection
1156899060 J * micah ~micah@micah.riseup.net
1156899571 J * Johnnie ~jdlewis@static-acs-24-154-32-33.zoominternet.net
1156899600 M * Bertl evening Johnnie!
1156900124 M * Johnnie Hey Bertl. :)
1156900611 J * shedi ~siggi@inferno.lhi.is
1156900796 M * Bertl hey shedi!
1156900801 Q * tanjix Server closed connection
1156900821 M * shedi thank you Bertl 
1156900822 J * tanjix ~tanjix@office.star-hosting.de
1156901615 J * Brio ~Brio@pool-68-238-62-224.port.east.verizon.net
1156901626 M * Bertl welcome Brio!
1156901630 M * Brio Hi
1156901683 M * Brio you know any website that contains the following data: Super Metroid (JU) [!].smc
1156901710 M * Bertl how's that related to Linux-VServer?
1156901729 M * Brio sorry
1156901755 M * Brio it's a ROM, for a emulator: ZSNES
1156901831 M * Bertl well, guess you have to search elsewhere for that ...
1156901875 M * Brio ah well, i might try google
1156901889 M * Brio thanks for the welcome n.n
1156902014 M * Bertl np, have fun and good luck!
1156902133 M * Brio thanks
1156903610 Q * Brio Quit: Brio
1156903856 J * shedii ~siggi@inferno.lhi.is
1156904028 Q * shedi Ping timeout: 480 seconds
1156904037 J * s4edi ~siggi@inferno.lhi.is
1156904338 Q * shedii Ping timeout: 480 seconds
1156904397 J * shedii ~siggi@inferno.lhi.is
1156904713 Q * s4edi Ping timeout: 480 seconds
1156904898 Q * shedii Ping timeout: 480 seconds
1156908721 J * sladen paul@starsky.19inch.net
1156911042 Q * wenchien Ping timeout: 480 seconds
1156911659 M * Bertl okay, I'm off to bed now .. have a good one everyone! cya!
1156911668 N * Bertl Bertl_zZ
1156911736 J * shedi ~siggi@inferno.lhi.is
1156913632 Q * cemil Ping timeout: 480 seconds
1156913646 Q * derjohn2 Ping timeout: 480 seconds
1156913657 Q * BenBen Ping timeout: 480 seconds
1156914060 J * derjohn2 ~aj@dslb-084-058-213-170.pools.arcor-ip.net
1156914599 Q * olilo Quit: bbiab
1156914635 J * olilo hiddenserv@tor.noreply.org
1156915888 J * cdrx ~legoater@cap31-3-82-227-199-249.fbx.proxad.net
1156917656 J * dna ~naucki@68-235-dsl.kielnet.net
1156918188 J * cemil ~cemil@defiant.wavecon.de
1156918221 J * BenBen ~benny@defiant.wavecon.de
1156921004 J * ybanafa ~admin@82.114.186.57
1156921044 Q * ybanafa 
1156922022 Q * mugwump Server closed connection
1156922027 J * mugwump ~samv@watts.utsl.gen.nz
1156922175 M * cdrx morning
1156922711 J * meandtheshell ~markus@85-124-36-216.dynamic.xdsl-line.inode.at
1156922977 N * kaner_ kaner
1156923192 J * id23 ~id@p5081474A.dip0.t-ipconnect.de
1156923401 J * ybanafa ~admin@82.114.186.57
1156923952 P * ybanafa Leaving
1156924407 J * coocoon ~coocoon@p54A054B9.dip.t-dialin.net
1156924480 J * ||Cobra|| ~cob@pc-csa01.science.uva.nl
1156924847 M * nayco_work 03:35 < Bertl : Well, adding .smc support Linux-VServer is definitely something to do :D !
1156925173 Q * derjohn2 Ping timeout: 480 seconds
1156925204 J * derjohn2 ~aj@dslb-084-058-213-170.pools.arcor-ip.net
1156926718 Q * id23 Remote host closed the connection
1156927557 M * nayco_work Hello, irc logs :D !
1156927574 M * nayco_work and hello all.
1156928372 Q * coocoon Quit: KVIrc 3.2.0 'Realia'
1156928419 A * Wonka throws some log after nayco_work 
1156928437 M * Wonka at
1156928445 M * Wonka damn germanisms
1156930247 J * coocoon ~coocoon@p54A0725D.dip.t-dialin.net
1156931098 Q * shedi Quit: Leaving
1156931972 M * nib-nico Wonka: German speak heavy speek?
1156932102 Q * michal_ Ping timeout: 480 seconds
1156932437 J * michal_ ~michal@www.rsbac.org
1156932752 J * SpongeBob ~spongebob@85-124-16-172.dynamic.xdsl-line.inode.at
1156932761 M * Wonka nib-nico: something like that
1156933149 M * nib-nico Question: in the dlimitsdirectory I have the file inodes_total. How can I get the number of Inodes I should use there?
1156933660 Q * ||Cobra|| Remote host closed the connection
1156934571 J * lilalinux ~plasma@dslb-084-058-206-238.pools.arcor-ip.net
1156935819 J * shedi ~siggi@dsl-149-109-85.hive.is
1156936562 J * ||Cobra|| ~cob@pc-csa01.science.uva.nl
1156936959 P * SpongeBob 
1156937246 Q * derjohn2 Remote host closed the connection
1156937276 Q * FaUl Ping timeout: 480 seconds
1156937441 M * SNy nib-nico: Not at all. Neither "speak" nor "heavy" would be words to use to properly express that phrase.
1156937711 M * nib-nico SNy: That was the joke...
1156937880 Q * cdrx Ping timeout: 480 seconds
1156937959 M * SNy Hehe.
1156938034 M * SNy Would be more appropriate to substitute English for German in that case. ;d
1156939142 Q * balbir Ping timeout: 480 seconds
1156941059 J * cdrx ~legoater@242.32.96-84.rev.gaoland.net
1156942819 Q * FireEgl Read error: Connection reset by peer
1156943323 J * FireEgl FireEgl@Sebastian.Atlantica.US
1156946170 N * Bertl_zZ Bertl
1156946174 M * Bertl morning folks!
1156946217 J * s0undt3ch_ ~s0undt3ch@bl7-242-127.dsl.telepac.pt
1156946260 M * Bertl nib-nico: the tools will use vdu to calculate the 'current' values, you only set the 'limits', assume one inode per dis and file
1156946265 M * ntrs Hi Bertl 
1156946295 M * Bertl hey ntrs!
1156946303 M * nayco_work 'llo, Bertl 
1156946369 M * Bertl hey nayco_work!
1156946497 Q * FireEgl Read error: Connection reset by peer
1156946662 Q * s0undt3ch Ping timeout: 480 seconds
1156946672 N * s0undt3ch_ s0undt3ch
1156946728 Q * ||Cobra|| Remote host closed the connection
1156947081 J * FireEgl FireEgl@Sebastian.Atlantica.US
1156947195 Q * sladen Ping timeout: 480 seconds
1156947405 J * sladen paul@starsky.19inch.net
1156949000 M * matti Hi Bertl.
1156949125 J * lilalinux_ ~plasma@dslb-084-058-239-065.pools.arcor-ip.net
1156949156 M * Bertl hey matti! :]
1156949555 Q * lilalinux Ping timeout: 480 seconds
1156950278 M * nib-nico thanks Bertl, that I'm setting only the limits is clear. Calculating one inode per file or dir is the thing haven't known. So if I don't care much about the inodes I can set the space-total limit to my wanted size and the the inodes only have to be a value which is big enough
1156950420 M * essobi_ Is there a way to limit a guest to the number of CPU threads it's allowed to consume, or a percentage of the total CPU available?
1156950455 M * Bertl nib-nico: well, the way a unix filesystem is built, makes the inode limit initeresting too
1156950455 M * derjohn essobi_, yes
1156950469 M * derjohn essobi_, look in the wiki for "scheduler"
1156950518 M * Bertl nib-nico: i.e. the total number of inodes is limited, and an evil guest could exhaust this limit without ever reaching the space limits
1156950527 M * essobi_ Bertl Reiser doesn't suffer from Inode limiting thou does it?
1156950550 M * Bertl essobi_: not directly, but indirectly too
1156950569 M * essobi_ It suffers from block size limiting I think..
1156950584 M * Bertl i.e. creating directory entries of size 0, causes metadata structures to grow, without accounting for space
1156950596 M * nib-nico Bertl: Ok, but for a good value I have to know an approximate value. And this value comes with some kind of experience I think.
1156950603 M * essobi_ Bertl Ah.
1156950627 M * Bertl nib-nico: well, there are simpler ways, e.g. 'df -i' will give you an idea of the filesystems themselves
1156950643 M * Bertl nib-nico: and 'vdu' will allow you to see the usage for a guest/dir
1156950659 M * essobi_ nib-nico hashing them would give you some consolidation to Inodes thou, yes?  Or does a link consume an inode as well?
1156950699 M * Bertl hard links do not consume inodes, only space in the dentry inode, soft links do consume inodes
1156950786 M * Bertl as a rule of thumb, you can say, 3 times the current is probably enough, and you should never give more than 1/3rd of the total resources
1156950789 M * nib-nico Bertl: OK, I will look for that to decide for some good values. I think if I want really evil guests to be limited I use an extra LVM partition
1156950820 M * Bertl nib-nico: that's a good idea in any case, but sometimes it's just an accident not pure evil behaviour :)
1156950842 M * Bertl i.e. consider a runaway script creating many, many empty files :)
1156950890 M * essobi_ while [ 0 ] ; do touch `date`.txt ; done 
1156950909 M * essobi_ there's 1 file per second..
1156950914 M * Bertl well, that would take some while, but yes
1156950914 M * nib-nico Bertl: full ACK
1156951026 M * essobi_ Anyone wrote some libs to process /proc/sys/vserver?  :)
1156951178 M * nib-nico what is the address of the LOG of this channel?
1156951185 M * essobi_ I was thinking of something to parse /proc/virtual/x/limit and cvirt into RRD/log files could be interesting..
1156951247 M * Bertl we are currently extending the syscall interface to gather this data via syscalls, but yes, feeding them into rrd databases is done for several years now
1156951252 M * nib-nico ah got it
1156951299 M * essobi_ Bertl Any examples anywhere, or is this all custom hacks?
1156951361 M * matti Hm...
1156951394 M * Bertl essobi_: I remember some munin plugin, but mostly rrdcollect is doing what you want
1156951562 M * essobi_ Bertl Neat.
1156951616 J * ntrs_ ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com
1156951623 Q * ntrs Read error: Connection reset by peer
1156951769 M * essobi_ hah.. the vserver script for munin isn't published.. it says their wiki says to ask for it in here. ;)
1156951878 M * Bertl well, guess you ahve to wait for _somebody_ to answer it then
1156951884 M * essobi_ ;)
1156951889 Q * shedi Quit: Leaving
1156952119 Q * FireEgl Quit: Bye...
1156952153 J * hallyn ~xa@ppp-69-217-236-19.dsl.klmzmi.ameritech.net
1156952398 M * Bertl wb hallyn!
1156952410 M * daniel_hozac essobi_: http://svn.debian.org/wsvn/pkg-vserver/people/holger/munin-plugins/?rev=0&sc=0
1156952454 M * essobi_ Nice.
1156953179 J * Viper0482 ~Viper0482@p54977575.dip.t-dialin.net
1156953204 J * FireEgl FireEgl@Sebastian.Atlantica.US
1156953425 Q * coocoon Quit: KVIrc 3.2.0 'Realia'
1156953722 Q * FireEgl Ping timeout: 480 seconds
1156953830 M * mnemoc hi, when should i mount something (post-start and friends) to get the entry on guests's mtab and not on the host?
1156953861 M * Bertl best to do it with the fstab in config, that should handle the mtab properly
1156953923 M * mnemoc my problem is that the mount point (/dev/pts) doesn'T exist because the parent (/dev) is tmpfs :\
1156954009 M * mnemoc i was thinking in mounting it with -n but looks evil
1156954085 M * daniel_hozac pre-start using secure-mount?
1156954116 M * Bertl or create the tmpfs first, prepare it an --move mount it
1156954140 M * mnemoc i'm mounting there, without secure-mount.... but it gets on the host.... does 'secure-mount' makes any trick in there?
1156954152 M * mnemoc Bertl: --move ? .oO
1156954169 M * daniel_hozac secure-mount has options to modify the guest's mtab, it's what util-vserver uses to mount the filesystems in fstab...
1156954218 M * mnemoc i thought secure-mount was a capability :p .... i'll take a look on both options
1156954305 M * essobi_ Nice.. Someone wrote a collectd plugin for vserver all ready.
1156954307 A * essobi_ apt-gets.
1156954422 M * daniel_hozac it's is a ccap too :)
1156954431 M * daniel_hozac (yes, kind of confusing...)
1156954606 J * FireEgl FireEgl@Sebastian.Atlantica.US
1156954616 M * mnemoc daniel_hozac: secure-mount at post-start place it on host's :( ..... i'll try at pre-start where i'm populating /dev
1156954634 M * daniel_hozac you'll have to supply the correct options.
1156954644 M * daniel_hozac it will default to the host's, but --mtab should tell it to look elsewhere.
1156954671 M * mnemoc ah, ok :)
1156954691 M * mnemoc the included help was not very complete on that topic :p
1156954739 M * daniel_hozac hmm... looks pretty complete to me.
1156954779 M * mnemoc daniel_hozac: probably i'm looking at the wrong place then
1156954798 M * mnemoc ah, found
1156954800 M * mnemoc sorry
1156954839 M * mnemoc i was just seeing "Invalid <src> <dst> pair specified"
1156955303 M * daniel_hozac ah.
1156956155 Q * cdrx Ping timeout: 480 seconds
1156956771 M * mnemoc daniel_hozac: thanks :) it works like a charm.... beside the fact than executing files from /usr/lib is not 'elegant' :p
1156956794 M * daniel_hozac yeah...
1156956809 M * daniel_hozac i guess you could source util-vserver-vars and use $_SECURE_MOUNT.
1156956827 M * mnemoc that sounds better
1156957292 Q * ntrs_ Quit: Leaving
1156957469 J * ntrs ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com
1156957667 J * mkhl mkhl@200-153-181-179.dsl.telesp.net.br
1156958587 J * shedi ~siggi@inferno.lhi.is
1156959511 M * nayco_work Bertl: I've got another question ;-) : What about Linux-VServer inclusion in the official kernel ? Is it still planned ?
1156959871 M * Bertl well, we are still working with the kernel folks, if you mean that
1156959926 M * Bertl but the 'framework' which will get into mainline will be something different, i.e. a common base for all the technologies around
1156960100 Q * sladen Ping timeout: 480 seconds
1156960110 J * sladen paul@starsky.19inch.net
1156960189 M * matti SELinux is by default in 2.6?
1156960296 M * nayco_work Bertl: ... You mean :
1156960324 M * nayco_work 1) A vanilla kernel won't be able to run vserver without patching anyway ?
1156960358 M * nayco_work 2) Some virtualisation technologies may share common code in near future ?
1156960419 M * nayco_work 3) Is this related to the new virtualisation-oriented instruction sets AMD and Intel introduced in their latest architectures ?
1156960451 M * Bertl well, mainline is _slowly_ considering virtualization technologies like Linux-VServer, first there will be some kind of 'namespace' solution, basically extending private (filesystem) namespaces to other areas like pid and networking
1156960482 M * Bertl regarding 1) at least not in the next 1-2 years, I guess
1156960497 M * Bertl ad 2) yes, I assume so :)
1156960525 M * Bertl ad 3) no, unrelated to those technologies, they are only relevant for xen and similar
1156960645 M * nayco_work Ok, so I won't talk about all this in the article ;-). Thanks, but could you develop a little for 2) ? You mean, different virt. projects' devs are talking about mergin code ?
1156960749 M * Bertl well, if you look at all the available projects, they share at least some functionality, which, when included in mainline, will be shared (to some extend)
1156960824 J * cdrx ~legoater@cap31-3-82-227-199-249.fbx.proxad.net
1156960833 M * Bertl welcome cdrx!
1156961013 M * nayco_work Ok. Do you thinks it is useful to talk about 3) in the article ? (I think it is too early)
1156961016 M * nayco_work -s
1156961074 M * Bertl as I said, those changes do not affect lightweight virtualization at all, they are only relevant for xen type (full) virtualizations
1156961079 M * nayco_work Sorry, not 3), but 2)
1156961087 M * nayco_work ;-)
1156961103 M * Bertl well, nothing shared yet, so not releavant ATM
1156961119 M * brc_ Good afternoonn! Is it possible to have a "includefile" in the place of the excludefile of vhashify ? I want just to include /usr /bin /sbin and not "everything but X Y Z directories".
1156961160 M * nayco_work Yes, for 3), I read something about this... I now remember it is mainly to run guest OSes without modifying them... Like some proprietary OS you know ;-)
1156961717 M * daniel_hozac brc_: it's an rsync-style exclude list, so that should be possible, IIRC.
1156961738 M * daniel_hozac i.e. /etc and +/etc/hostname would exclude all of /etc, except /etc/hostname.
1156962222 J * bonbons ~bonbons@83.222.36.236
1156962407 M * daniel_hozac Bertl: any plans to include the COW breaking on chmod/chown/etc. in 2.1.1?
1156962432 M * brc_ i tried that btu did not work. maybe cause i was using --manually. going to try again.
1156962785 M * Bertl daniel_hozac: I guess we can do that ..
1156962840 M * mnemoc daniel_hozac: funny thing, with secure_mount i'm getting /dev/pts as nodev
1156962867 M * daniel_hozac mnemoc: yeah, it defaults to nodev, i think you need to explicitly specify -o dev.
1156963010 M * Bertl daniel_hozac: currently I'm trying to figure out the issues ntrs is (still) encountering with devel vs. stable (and mysql)
1156963041 M * daniel_hozac so we're certain it's a stable vs. devel issue now? i.e. 2.6.17.11-vs2.0.2-rc31 got tested too?
1156963049 M * Bertl daniel_hozac: disabling the COW and HARD scheduler dud not change much there, njow looking for more difficile changes
1156963070 M * Bertl yes, it is not present with 2.0.2-rc31
1156963091 Q * mkhl 
1156963104 M * daniel_hozac ok, that's weird then.
1156963282 M * Bertl yes, indeed, disabling COW+HARD changed the way the issues show up slightly
1156963311 M * Bertl in the 'normal' setyp, ntrs ends up with many, many locked tables, and zero progress
1156963345 M * Bertl with that disabled mysql locks up too, but didn't show the locked tables (only one test so far)
1156963460 M * daniel_hozac sounds very strange... how does mysql do the locking?
1156963606 M * Bertl no idea, but it seems related to NPTL or TLS usage, it favors FC guests and seems to go away with single threaded mysqld
1156963809 Q * FireEgl Read error: Connection reset by peer
1156963992 Q * lilalinux_ Remote host closed the connection
1156964410 J * FireEgl FireEgl@Sebastian.Atlantica.US
1156965356 Q * brc_ Quit: BitchX-1.1-final © 1996-2000 Colten Edwards
1156965486 Q * Viper0482 Remote host closed the connection
1156966912 J * PowerKe ~icuser@d5153A0DA.access.telenet.be
1156966947 J * ekc ~ekc@c-24-23-195-211.hsd1.ca.comcast.net
1156966949 Q * hallyn Ping timeout: 480 seconds
1156967259 Q * FloodServ iridium.oftc.net helium.oftc.net
1156967262 J * FloodServ services@services.oftc.net
1156967267 Q * Nam iridium.oftc.net xenon.oftc.net
1156967267 Q * click iridium.oftc.net xenon.oftc.net
1156967267 Q * cehteh iridium.oftc.net xenon.oftc.net
1156967267 Q * Greek0 iridium.oftc.net xenon.oftc.net
1156967267 Q * hap iridium.oftc.net xenon.oftc.net
1156967267 Q * [PUPPETS]Gonzo iridium.oftc.net xenon.oftc.net
1156967267 Q * bogus iridium.oftc.net xenon.oftc.net
1156967267 Q * tokkee iridium.oftc.net xenon.oftc.net
1156967267 Q * mcp iridium.oftc.net xenon.oftc.net
1156967267 Q * nokoya iridium.oftc.net xenon.oftc.net
1156967277 Q * ntrs iridium.oftc.net nova.oftc.net
1156967277 Q * michal_ iridium.oftc.net nova.oftc.net
1156967287 J * ntrs ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com
1156967287 J * michal_ ~michal@www.rsbac.org
1156967303 J * Nam ~nam@S0106001195551ff0.va.shawcable.net
1156967303 J * click click@ti511110a080-2980.bb.online.no
1156967303 J * cehteh ~ct@cehteh.homeunix.org
1156967303 J * Greek0 ~greek0@85.255.145.201
1156967303 J * bogus ~bogusano@fengor.net
1156967303 J * nokoya young@hi-230-82.tm.net.org.my
1156967303 J * mcp ~hightower@wolk-project.de
1156967303 J * tokkee tokkee@casella.verplant.org
1156967303 J * hap ~penso@212.27.33.226
1156967303 J * [PUPPETS]Gonzo gonzo@langweiligneutral.deswahnsinns.de
1156967503 M * matti http://deadbeef.info/tmp/patch-2.6.17.11-grsec2.1.9-ipv1-vs2.0.2-rc31.diff.gz
1156967504 M * matti ;]
1156967621 Q * Zaki Remote host closed the connection
1156967642 M * Bertl daniel_hozac: it seems that a devel kernel with the following patch (revers) is now running fine : http://vserver.13thfloor.at/Stuff/delta-rc31.2r.diff
1156967680 M * Bertl but we are not 100% positive about that, because it first did hit a process limit, and this could have been the reason for all the issues in the first place
1156967729 M * Bertl so we are going to keep that kernel running for a few hours, to confirm that this kernel is fine, then we'll switch back to an unmodified 2.1.1-rc31
1156967760 M * Bertl if that exposes the same issues, without hitting any limits, we have a positive
1156967788 M * Bertl in the meantime, I will medidate about possible unwanted sideeffects of those changes
1156967815 M * Bertl (any ideas are welcome :)
1156967879 M * Bertl ntrs can give more details about the setup, if required
1156968302 J * ntrs_ ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com
1156968302 Q * ntrs Read error: Connection reset by peer
1156968457 J * yarihm ~yarihm@84-75-123-221.dclient.hispeed.ch
1156968662 J * Zaki ~Zaki@212.118.126.43
1156969143 Q * ekc Ping timeout: 480 seconds
1156969215 J * ntrs__ ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com
1156969215 Q * ntrs_ Read error: Connection reset by peer
1156969222 M * daniel_hozac hmm, that's all it takes to get it working again?
1156969317 J * Hollow_Berlin ~bene@217.110.45.98
1156969323 A * Hollow_Berlin waves
1156969344 A * Bertl waves back
1156969351 M * Hollow_Berlin :)
1156969365 M * mnemoc :o
1156969375 J * hallyn ~xa@ppp-69-217-236-19.dsl.klmzmi.ameritech.net
1156969376 M * daniel_hozac how's the hunt going?
1156969379 M * Hollow_Berlin sony center hotspot works great
1156969404 M * Hollow_Berlin well, falt hunting is tomorrow... we applied for our studies today...
1156969409 M * Hollow_Berlin *flat
1156969413 M * daniel_hozac ah, ok.
1156969427 M * mnemoc Hollow: will you come to live in Berlin?
1156969433 M * Hollow_Berlin yep :)
1156969436 M * mnemoc :D
1156969513 M * daniel_hozac Bertl: any particular reason we don't use xchg for both xid and vxi?
1156969633 M * daniel_hozac or does xchg only work on longs?
1156969666 Q * Hollow_Berlin Quit: This computer has gone to sleep
1156969751 M * Bertl daniel_hozac: don't remember, it had a reason ...
1156969784 M * daniel_hozac ok.
1156970092 Q * bonbons Quit: Leaving
1156970480 J * Hollow_Berlin ~bene@217.110.45.98
1156970491 M * Hollow_Berlin low on battery :(
1156970516 M * Hollow_Berlin some photos of the garden inside the informatics institute of my uni: http://home.xnull.de/work/berlin/ :)
1156970568 M * derjohn Hollow_Berlin, hey, Berlin ;)
1156970629 M * derjohn Do they grow dope inside the faculty? :-o
1156970868 N * ntrs__ ntrs
1156971020 Q * Hollow_Berlin Ping timeout: 480 seconds
1156971184 M * matti Hm, tree in a middle of building?;]
1156971191 M * matti Kinda kinky ;]
1156972740 M * Bertl daniel_hozac: okay, it seems confirmed, devel exposes the issues
1156972770 M * Bertl we will continue bisection tomorrow
1156972881 J * Aiken ~james@tooax6-090.dialup.optusnet.com.au
1156972920 M * Bertl morning Aiken!
1156973031 M * Aiken hello
1156973975 Q * BenBen Quit: Lost terminal
1156973975 Q * cemil Quit: Lost terminal
1156974966 Q * yarihm Quit: Leaving
1156975594 Q * meandtheshell Quit: bye bye ... 
1156975702 N * Nam Nambrb
1156975734 J * nammie ~nam@S0106001195551ff0.va.shawcable.net
1156975790 N * nammie Nam
1156976187 Q * Nambrb Ping timeout: 480 seconds
1156976346 J * yarihm ~yarihm@84-75-123-221.dclient.hispeed.ch
1156977622 Q * dna Quit: Verlassend
1156977857 Q * hallyn Quit: leaving
1156978333 M * daniel_hozac Bertl: even with those hunks reverted?
1156978420 Q * yarihm Quit: Leaving
1156978550 M * Bertl daniel_hozac: hmm?
1156978570 M * daniel_hozac 23:19 < Bertl> daniel_hozac: okay, it seems confirmed, devel exposes the issues
1156978599 M * daniel_hozac -> issues occur with http://vserver.13thfloor.at/Stuff/delta-rc31.2r.diff too? or not?
1156978602 M * Bertl yes, with the 3 hunks applied (i.e. changes removed) it worked fine for more than two hours
1156978622 M * Bertl we decided to double check the devel case then
1156978628 M * Bertl and it instantly hung
1156978652 M * Bertl so IMHO the issue is in one of those three changes
1156978665 M * daniel_hozac very odd...
1156978683 M * Bertl haven't found anything I could point at yet
1156978710 M * daniel_hozac will you be adding them back one by one to see which one is the culprit?
1156978720 M * Bertl some weird theory of mine is that the signalling gets wrong (realtime and timer)
1156978739 M * Bertl and yes, we will add them back one by one tomorrow
1156978750 M * Bertl (that's what I meant with bisection)
1156978786 M * daniel_hozac i figured, but wanted to make sure.
1156978885 M * daniel_hozac is the put_vx_info really needed? shouldn't leave handle that?
1156978996 M * Bertl no the leave/enter functions do not do any refcounting
1156979016 M * daniel_hozac ah, it's the task_vx_info_get, ok.
1156979041 M * daniel_hozac s/task_vx_info_get/task_get_vx_info/
1156979072 M * Bertl I also doubt that we are experiencing some kind of context goes away issues or such
1156979096 M * Bertl (guest was running ~300 thread, ~80 tasks)
1156979117 M * daniel_hozac yeah, it probably wouldn't cause that sort of issues either.
1156979152 M * Bertl I would opt for some lost timer event or something like that
1156979300 M * Bertl but the code looks perfectly fine and the changes are quite simple
1156979313 M * daniel_hozac i agree.
1156979592 M * matti Lalala.
1156979605 M * matti Hi daniel_hozac :)
1156979622 M * daniel_hozac hello matti
1156981767 Q * MrX Ping timeout: 480 seconds
1156982316 M * Bertl okay, I'm off to bed ... have a good one everyone! cya!
1156982322 N * Bertl Bertl_zZ