1162512527 J * lilalinux__ ~plasma@dslb-084-058-213-254.pools.arcor-ip.net
1162512920 M * Wonka SuSE is sometimes known as "the Nurembergian Windows"
1162512957 Q * lilalinux_ Ping timeout: 480 seconds
1162512961 A * Wonka likes debian way more
1162512981 M * Bertl ah, another round of DistroWars(tm)
1162513001 M * doener do we need dice?
1162513020 A * doener throws a few dice into the channel just in case
1162513022 M * Bertl nah, we have /dev/random :)
1162513042 M * doener I'll not throw my /dev/random in here ;)
1162513064 M * Bertl worried that you might expose secret data from the entropy pool, eh? *G*
1162513105 M * doener nah, more worried about me being lucky and getting a \n/quit I'm dumbshit\n out of /dev/random ;)
1162513116 Q * meandtheshell Quit: exit (0);
1162513145 M * daniel_hozac when /dev/random starts communicating, you've had too much of some more or less illegal substance ;)
1162513171 M * doener I'm going to a whisky tasting tomorrow, is that enough? ;)
1162513195 M * daniel_hozac hehe
1162513271 M * Bertl well, consider it, the chance that /dev/random will result in the complete work of W. Shakespeare is identical to the result you get when doing a dd if=/dev/random of=ws.tar.bz2 bs=1M count=60
1162513320 M * Bertl or the same as dd if=/dev/zero of=zero.dat bs=1M count=60 :)
1162513357 M * Bertl granted, chances are not too good, but hey, maybe you are lucky :)
1162513383 M * doener guess using the monkeys is a faster way to achieve that
1162513396 M * Bertl well, it's definitely more fun :)
1162513419 M * doener and you need to answer a lot of questions to get the required amount of bananas
1162513463 M * doener anyway, back to factoring numbers
1162513617 M * Bertl large numbers?
1162513640 M * doener I've already found at least 5 (at first sight) totally different ways to show that $odd-number = $x^2 - $y^2 where $x > $y
1162513656 M * doener doesn't help in any way though ;)
1162513690 M * Bertl hum, care to explain what you are trying to accomplish?
1162513721 M * doener nothing ;) It's just a way to relax and force my brain into thinking around corners
1162513749 M * doener the goal would be fast factoring of large numbers, but I won't accomplish that anyway (at least I'm pretty sure of that)
1162513755 M * Bertl i.c., well, when you're at it, could you verify that 2^49152-1 is prime?
1162513793 M * doener hm, mersenne prime? such a small one should already have been proven, shouldn't it?
1162513829 M * doener or do I mess up exponent with the actual number WRT found mersenne primes?
1162513845 M * doener no, I'm not
1162513863 M * Bertl nah, it's fine, I didn't want you to solve unsolveable stuff
1162513898 M * doener well, unless "fast" is a requirement, it's easily solvable ;)
1162513926 M * Bertl my last status is 2^32582657-1
1162513960 M * doener the point for which I've chosen that task is that I can probably keep on trying forever and try to find new ways. If I would always have new tasks, I would easily fall back into old habits
1162513976 Q * bronson Read error: Connection reset by peer
1162514068 J * bronson ~bronson@adsl-64-161-106-11.dsl.snfc21.pacbell.net
1162514617 M * derjohn2 if I "vserver foo stop" and "vserver foo start" suddenly the lvm partition/xfs /var/lib/vservers/foo gets read only. 
1162514667 M * daniel_hozac does foo have a lot of capabilities and run too many initscripts?
1162514720 M * derjohn2 I didnt add any capas, but left in the usual debian inits scripts (incl. the umount stuff at shutdown)
1162514868 M * derjohn2 hm, is there any useful differnce in fstab vs. fstab.remote ?
1162514895 M * derjohn2 /usr/lib/util-vserver/vserver.stop: line 85: 13039 Killed                  "${NICE_CMD[@]}" ${USE_VNAMESPACE:+$_VNAMESPACE --enter "$S_CONTEXT" -- } $_VCONTEXT $SILENT_OPT --migrate --chroot --xid "$S_CONTEXT" -- "${INITCMD_STOP[@]}"
1162514906 M * derjohn2 and that friend fires at shutdown
1162514942 M * daniel_hozac yes, fstab.remote is executed in the guest's network context.
1162514968 M * derjohn2 hm. but my network mounts did work in the fstab, too
1162515183 A * hardwire wonders wtf ubuntu is so not vserverish
1162515188 M * derjohn2 smbfs: Unrecognized mount option nodev
1162515206 M * derjohn2 hardwire, because they tended to integrate Xen first.
1162515213 M * daniel_hozac 0.30.211?
1162515232 M * derjohn2 daniel_hozac, dunno .. i just dist.upgraded 
1162515235 M * derjohn2 30 secs ago
1162515245 M * derjohn2 and added a fstab.remote now
1162515254 M * derjohn2 vxW: !!! limit: ffff81013ec93070[<NULL>,20] = 1 on exit.
1162515254 M * derjohn2 vxW: !!! limit: ffff81013ec93070[<NULL>,21] = 1 on exit
1162515279 M * derjohn2 all that is on rc46 .. and now I try with 311
1162515279 M * daniel_hozac so for some reason your semaphores aren't getting removed on shutdown.
1162515321 M * derjohn2 maybe because the unmount of the remote fs doesnt work?
1162515353 Q * DreamerC_ Quit: leaving
1162515357 M * daniel_hozac why doesn't it work?
1162515367 M * Bertl derjohn2: that's with rc46, I presume?
1162515371 J * DreamerC ~dreamerc@61-217-226-119.dynamic.hinet.net
1162515375 M * daniel_hozac but no, that's IPC.
1162515385 M * derjohn2 ehhh ... lart me! I assumed that i might not have worked
1162515395 M * derjohn2 Bertl, yes, rc46 on amd64
1162515410 M * daniel_hozac what are you running in that guest?
1162515417 M * derjohn2 ubuntu dapper
1162515423 M * derjohn2 host is debian etch
1162515423 M * hardwire derjohn2: yeh.. I guess I just enjoyed the deb testing/unstable packages for vserver
1162515425 M * daniel_hozac i meant more along the lines of which applications ;)
1162515430 Q * Aiken_ Quit: Leaving
1162515446 M * hardwire I was hoping to have an ubuntu edgy host
1162515458 M * hardwire with all binary kernels .. for support reasons
1162515466 M * hardwire and they locked it w/o any vserver kernels
1162515469 M * hardwire kinda torques me off
1162515487 M * daniel_hozac derjohn2: i think you should be able to list the semaphores with cat /proc/sysvipc/sem when you're in the guest (or from xid 1)
1162515492 M * hardwire its like.. right there.. right there just almost right I can reach it and then BOOM a window crushes your fingers
1162515493 Q * bonbons Remote host closed the connection
1162515524 M * daniel_hozac hardwire: can't you use the Debian kernels on Ubuntu?
1162515529 M * derjohn2 daniel_hozac, apps: apache2 ... not much more 
1162515533 M * hardwire I assume so
1162515544 M * hardwire daniel_hozac: I was just being uber wishfull
1162515564 M * daniel_hozac derjohn2: ah, well, httpd has caused similar problems in the past, IIRC.
1162515568 M * hardwire In reality I am just being impatient for etch to get stamped "stable"
1162515569 M * derjohn2 hardwire, daniel_hozac : dint try with md /softtaid. since edgy ubuntu seems to be incompatible with debian.
1162515576 M * hardwire so that I can justify deploying it on my machines
1162515598 M * hardwire yuck
1162515599 M * daniel_hozac i was under the impression etch was frozen?
1162515601 M * hardwire see thats why
1162515608 M * hardwire daniel_hozac: etch is not frozen
1162515616 M * hardwire I get multitudes of updates
1162515627 M * daniel_hozac i was told it froze sometime mid-october, IIRC :)
1162515636 M * derjohn2 hardwire, only updates for grave bugs
1162515640 M * hardwire well they need to fix the refer.. cause its thawing a bit
1162515693 M * hardwire interesting
1162515694 M * hardwire well
1162515699 M * hardwire as long as random strangers say so
1162515701 M * hardwire I will start deploying it
1162515760 M * daniel_hozac derjohn2: is your httpd dying by SIGKILL?
1162515773 M * daniel_hozac derjohn2: i think that's how we got the semaphores to stick around before.
1162515790 M * daniel_hozac Bertl: oh, btw, futex isolation, we'll do that for 2.1.2?
1162515805 M * Bertl yep, of course
1162515819 M * daniel_hozac ok, nice.
1162515861 M * derjohn2 daniel_hozac, possible .. killed by sending a SIGKILL signal. The following process ...
1162515869 M * derjohn2    \_ /usr/sbin/apache2 -t ..
1162515871 M * derjohn2 yes
1162515884 M * daniel_hozac derjohn2: so that's why you get those warnings then ;)
1162515929 M * Bertl I think we should implement some kind of zap-context at some point
1162515944 M * daniel_hozac ?
1162515945 M * derjohn2 hmm ... but why ? can I look with vserver --debug ? oder better ways ?
1162515964 M * Bertl daniel_hozac: well, releasing certain resources belonging to a context (on request)
1162515974 M * daniel_hozac derjohn2: why what? why it's killed by SIGKILL? because it took longer than 30 seconds to stop the guest.
1162515998 M * daniel_hozac Bertl: shouldn't all those resources be freed when it's stopped?
1162516003 M * daniel_hozac ideally, i mean.
1162516027 M * Bertl well, no, for example the futexes or certain shmem setups will not
1162516040 M * daniel_hozac right, but they should, no?
1162516050 M * daniel_hozac IMHO it's a bug to have locks around after restarting a guest.
1162516079 M * Bertl phew, tough question .. as they are not bound to tasks, they exist as long as the machine is up
1162516107 M * daniel_hozac but don't we tag the IPC stuff already?
1162516110 M * Bertl now the question is, when does a context actually shut down, and what resources do we want to cancel at this point
1162516147 M * Bertl yes, it's all tagged, and that's why I think that we should add a zap sooner or later :)
1162516150 M * daniel_hozac free_vx_info? and i'd say all resources.
1162516163 M * Bertl hehe, that'd be tricky
1162516179 M * daniel_hozac well, obviously excluding files cached and such.
1162516181 M * Bertl just assume resource X holds a reference to vx_info :)
1162516202 M * Bertl -> no context will ever get freed :)
1162516209 M * daniel_hozac what resources hold references?
1162516216 M * daniel_hozac and wouldn't that already be a problem in that case? ;)
1162516217 M * Bertl curently the sockets do
1162516233 M * daniel_hozac but sockets are cleaned when the processes die.
1162516240 M * Bertl some of them
1162516244 M * daniel_hozac oh?
1162516248 M * daniel_hozac TIME_WAIT?
1162516251 M * Bertl for example
1162516255 M * daniel_hozac ok.
1162516277 M * Bertl so, I think it's better to have an explicit interface to remove resources
1162516304 M * Bertl you can then decide to zap them on restart too, if you like
1162516327 M * daniel_hozac would probably be best to zap them on stop, which inludes restart ;)
1162516363 M * daniel_hozac but what more-persistent-than-processes resources are there?
1162516369 M * daniel_hozac sockets, IPC, futex, ...?
1162516421 M * Bertl maybe dentries (in the future), caches, ... 
1162516459 M * Bertl network mounts?
1162516469 M * daniel_hozac don't they expire with the namespace?
1162516484 M * Bertl well, namespaces can stick around too
1162516495 M * daniel_hozac hmm, even without a process in them?
1162516510 M * Bertl maybe even eliminate zombies from a context, to allow guest restart?
1162516533 M * Bertl (they could be moved to another context for example)
1162516540 M * daniel_hozac are there instances where zombies would keep a guest from restarting?
1162516548 M * daniel_hozac shouldn't the child_reaper handle those?
1162516562 M * Bertl sure, they will not allow a context to disappear if not reaped
1162516589 M * Bertl or what about processes in 'D' state?
1162516598 M * Bertl (because of a kernel bug, e.g.)
1162516601 M * daniel_hozac yeah, those would be handy.
1162516606 M * daniel_hozac (to be able to get rid of)
1162516714 M * Bertl we could have a purgatory context :)
1162516728 M * daniel_hozac hehe.
1162516768 M * Bertl or whatever the political correct term will be :)
1162516993 J * Aiken ~james@tooax6-009.dialup.optusnet.com.au
1162519159 Q * ntrs Quit: Leaving
1162519160 M * derjohn2 slowly I'm beginning to lose faith... but now after reboote the guests dont hang on shutdown anympore. 211 utils, smbfs, rc46 .... I neben mount the smb as user "nobody"
1162519814 M * cehteh mhm
1162520296 M * Bertl http://paste.linux-vserver.org/617
1162520308 M * Bertl okay, let me know if I forgot something there ... (2.1.1)
1162520446 M * Bertl I'm currently updating the debug.txt info and doing the absolutely final cleanup
1162520502 M * Bertl i.e. there will be an rc47, which will be released as 2.1.1 unmodified, I hope
1162522504 M * doener hope dies last, eh?
1162522510 M * doener ;)
1162522592 Q * ensc Killed (NickServ (GHOST command used by ensc_))
1162522603 J * ensc ~irc-ensc@p54B4D3E6.dip.t-dialin.net
1162522651 M * Bertl doener: right :)
1162525560 Q * bronson Ping timeout: 480 seconds
1162528566 J * dd ~mainius@c-24-3-25-33.hsd1.pa.comcast.net
1162528786 M * dd hello?
1162528879 Q * dd autokilled: Take your warez somewhere else. If you feel an error has been made, please contact support@oftc.net. (2006/11/2 23.41)
1162528886 Q * ensc Ping timeout: 480 seconds
1162528889 J * dd ~mainius@c-24-3-25-33.hsd1.mn.comcast.net
1162528889 Q * dd autokilled: Take your warez somewhere else. If you feel an error has been made, please contact support@oftc.net. (2006/11/2 23.41)
1162530042 J * bronson ~bronson@c-71-198-75-160.hsd1.ca.comcast.net
1162530870 M * Bertl okay, off to bed now ... debug.txt still needs some fixing ... will do that later today ...
1162530878 M * Bertl have a good one everyone! cya!
1162530883 N * Bertl Bertl_zZ
1162532307 Q * Vudumen Ping timeout: 480 seconds
1162532483 J * Vudumen ~vudumen@perverz.hu
1162533288 Q * mire_ Ping timeout: 480 seconds
1162534299 J * mire_ ~mire@116-167-222-85.adsl.verat.net
1162534489 Q * bronson Read error: Operation timed out
1162534501 J * bronson ~bronson@adsl-64-161-106-11.dsl.snfc21.pacbell.net
1162534883 Q * Aiken Quit: Leaving
1162537157 J * Piet_ hiddenserv@tor.noreply.org
1162537169 J * maverill ~forty_6_a@adsl-75-5-231-54.dsl.klmzmi.sbcglobal.net
1162537181 P * maverill 
1162537240 J * paiman ~qwerty@202.46.129.84
1162537300 J * maverill ~forty_6_a@adsl-75-5-231-54.dsl.klmzmi.sbcglobal.net
1162537313 P * maverill 
1162537563 Q * Piet Ping timeout: 480 seconds
1162537703 Q * paiman 
1162538622 J * maverill ~forty_6_a@adsl-75-5-231-54.dsl.klmzmi.sbcglobal.net
1162538627 P * maverill 
1162539045 J * meandtheshell ~markus@85-124-232-81.work.xdsl-line.inode.at
1162539216 Q * mountie Ping timeout: 480 seconds
1162539222 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com
1162540888 Q * MrX Ping timeout: 480 seconds
1162540933 Q * Piet_ Ping timeout: 480 seconds
1162541167 J * Piet_ hiddenserv@tor.noreply.org
1162541404 M * nayco_work Hello, all !
1162542209 J * ensc ~irc-ensc@p54B4D529.dip.t-dialin.net
1162542304 J * trippeh_ atomt@x.vx.no
1162542353 J * MrX ~urk@60.49.42.241
1162542423 Q * trippeh Ping timeout: 480 seconds
1162542709 J * dna_ ~naucki@91-236-dsl.kielnet.net
1162543476 M * renihs hello nayco_work 
1162543814 M * nayco_work Ok, I gonna try with -rc46
1162543880 J * sebastian ~sebastian@pD957DCA3.dip.t-dialin.net
1162545452 Q * shedi Quit: Leaving
1162545536 Q * kugg Ping timeout: 480 seconds
1162546803 J * Aiken ~james@tooax6-009.dialup.optusnet.com.au
1162546916 J * FireEgl FireEgl@Sebastian.Atlantica.US
1162547604 J * bonbons ~bonbons@83.222.36.111
1162548090 J * kugg kugg@illvilja.org
1162549346 Q * ensc Ping timeout: 480 seconds
1162549671 Q * Zaki Ping timeout: 480 seconds
1162551337 J * ensc ~irc-ensc@p54B4DBEB.dip.t-dialin.net
1162552372 J * kir_home ~kir@213.152.157.70
1162552942 Q * ruskie Remote host closed the connection
1162553263 J * ruskie ~ruskie@ruskie.user.oftc.net
1162553354 Q * ruskie 
1162553548 J * ruskie ~ruskie@ruskie.user.oftc.net
1162554377 Q * FireEgl Quit: Bye...
1162554631 Q * Aiken Ping timeout: 480 seconds
1162555324 J * dna___ ~naucki@3-222-dsl.kielnet.net
1162555617 J * wam ~wam@proxy1.msh.de
1162555628 M * wam Hi, how do I add a ip-address to a running vserver?
1162555643 M * daniel_hozac does the guest have more than one IP address already?
1162555647 M * wam yes
1162555664 M * daniel_hozac vcmd should be able to do it then.
1162555681 Q * dna_ Read error: Operation timed out
1162555684 M * wam vcmd?
1162555709 M * wam should this be in utils-vserver?
1162555725 M * daniel_hozac ./vcmd -i <nid> -C net_add .type=1 .count=1 .ip[0]=<IP> .mask[0]=<mask>
1162555734 M * daniel_hozac no, http://vserver.13thfloor.at/Experimental/TOOLS/vcmd-0.06.tar.bz2
1162555755 M * wam cool - i'll get this. thanks!
1162555769 M * wam btw: will this work with vserver 2.0.2-rc10 ?
1162555780 M * daniel_hozac it should.
1162555788 M * daniel_hozac but why are you running such an old kernel?
1162555801 M * wam because I can't reboot that machine every half a year ;)
1162555891 M * daniel_hozac 2.0.2-rc10 is a bad version to stabilize on though, there have been two security issues since ;)
1162555933 M * wam hmm... I just hate downtime ;)
1162555960 M * wam is <nid> the id of the context?
1162555976 M * daniel_hozac if you're using util-vserver 0.30.210+, yes.
1162556033 M * wam Will I have to up the IP before the command?
1162556046 M * daniel_hozac yes.
1162556180 M * wam is [0] the index of the context's IP? Won't I overwrite the currently first ip?
1162556190 M * daniel_hozac no.
1162556211 M * daniel_hozac the [0] is because you can add up to 4 IP addresses at the same time.
1162556592 Q * Piet_ Quit: Piet_
1162557352 M * wam ok - I'm just a keypress away from adding the ip. One more thing. I use the alias "eth0:squid" for the interface. Won't I have to submit that name or at least the interface? - A "no" as answer will be enough ;)
1162557436 M * daniel_hozac no.
1162557488 M * daniel_hozac networking is entirely IP-based, interfaces don't matter at all.
1162557581 Q * kir_home Ping timeout: 480 seconds
1162557780 M * renihs hmm anyone using mrtg/snmp or alike here to monitor vserver guests? 
1162557804 M * renihs i currently graph the host system, however it would be nice to see at a glance which guest is messing around
1162557826 M * daniel_hozac i think a lot of people are using collectd or munin to graph guests.
1162557843 M * daniel_hozac (vserver plugins available for both)
1162557850 M * renihs munin? hmm i know neither
1162557853 M * daniel_hozac but they monitor from the host.
1162557859 M * renihs that doesnt matter
1162557934 M * daniel_hozac http://collectd.org/plugins/vserver.shtml
1162558562 M * renihs hmm just checked munin, but i am confused
1162558566 M * renihs lemme check that url
1162558606 M * renihs that one looks like what i want :)
1162559118 M * h01ger micah, gdm: http://collectd.org/plugins/vserver.shtml#memory_usage is a good default view for single vserver memory consumption
1162559218 M * h01ger micah, gdm: and the new munin package supports building of a munin-plugins-contrib package (disabled per default), btw. and i'm too busy with other stuff to do something in this area :(
1162559231 M * gdm h01ger: that is essentially the view i had with the vsmem_ graphs i was using, except solid rather than line graphs
1162559237 M * h01ger yup
1162559252 M * h01ger but its gone atm 
1162559293 M * gdm yeah, i'm too busy too
1162559308 M * gdm i haven't really done anything with any of the plugins for ages in fact
1162560568 M * Hollow derjohn: cheater!
1162560570 M * Hollow ;)
1162560669 M * derjohn Hollow, "forum l8mer" ;) !
1162560704 M * Hollow and you still have to apply to "men vorname ist ein nachname" ;)
1162560705 M * renihs narf
1162560737 M * derjohn Hollow, does that really exist? If not, I will create it ;)
1162560743 M * Hollow it does .. ;)
1162561048 P * marcfiu 
1162561097 Q * adrien Ping timeout: 480 seconds
1162561403 J * marcfiu ~mef@targe.CS.Princeton.EDU
1162561406 M * marcfiu hello
1162561577 M * renihs hmm great, i have never used rrdtool before, any quick way creating a graph from a .rrd file?
1162561743 M * daniel_hozac marcfiu: how's the patching coming along?
1162561900 M * daniel_hozac and btw, remember to remove the put_task_struct from find_proc_task_by_pid, or bad things will happen ;)
1162561991 M * derjohn Hollow: done :)
1162562020 M * Hollow hehe
1162562035 M * derjohn Hollow: We should Bertl_zZ tell to apply there ;)
1162562048 M * derjohn (of course only for the Linux-VServer forum!)
1162562079 M * Hollow lol..
1162562120 M * daniel_hozac hmm?
1162562224 M * derjohn daniel_hozac, we talk about a german student's diretory. by random i found Hollow there ...
1162562239 M * daniel_hozac ah.
1162562254 M * derjohn daniel_hozac, http://www.studivz.net <-- you can apply there too, but I doubt that they know about your university. 
1162562272 M * daniel_hozac hehe, i'm pretty sure they don't ;)
1162562290 M * daniel_hozac (and my german is really terrible...)
1162562298 M * derjohn daniel_hozac, but maybe ... actually German students tend to spend a semester in a foreign country .. so it may be listed. or simply cheat :)
1162562300 J * _tanjix ~tanjix@office.star-hosting.de
1162562313 M * Hollow well, studivz rather sucks
1162562321 M * derjohn daniel_hozac, you speak German ?
1162562329 M * daniel_hozac not really, no ;)
1162562342 Q * SNy Remote host closed the connection
1162562344 J * SNy 7991fe9e90@bmx-chemnitz.de
1162562352 M * daniel_hozac i'm pretty much limited to saying what my name is and where i live ;)
1162562354 Q * tanjix Read error: Connection reset by peer
1162562359 N * _tanjix tanjix
1162562385 M * derjohn Hollow, yes, the technology is overloaded, but it only there for "gruscheln" , nor?
1162562416 M * derjohn i think my name in swedish is "Andreas John" ;)
1162562418 M * Hollow lol.. gruscheln is the most useless thing i've ever seen in a community ..
1162562464 M * derjohn Hollow, that why we like it :-)
1162562538 J * s0undt3ch_ ~s0undt3ch@bl4-62-140.dsl.telepac.pt
1162562965 Q * s0undt3ch Ping timeout: 480 seconds
1162562965 N * s0undt3ch_ s0undt3ch
1162563035 M * Hollow daniel_hozac: please take a look at http://people.linux-vserver.org/~hollow/util-vserver/delta-gentoo-vsomething-feat01.diff again.. i fixed the yum thing, and added vesync (the last missing tool)
1162563053 M * Hollow ah
1162563061 M * Hollow and it includes a gentoo init style feature ;)
1162563092 M * Hollow wait ..
1162563097 M * Hollow forget to svn add vesync
1162563139 M * daniel_hozac isn't vdispatch-conf missing "$@"? or doesn't it take arguments?
1162563141 M * Hollow should be ok now
1162563146 M * Hollow it doesn't take any
1162563170 M * daniel_hozac at all? heh, strange script.
1162563185 M * Hollow yup, really none.. i double-checked ;)
1162563208 M * Hollow regarding vesync we don't want it to take any other
1162563314 M * Hollow probably we should add --nocolor to emerge in vesync, so it works with mails from cron..
1162563453 M * daniel_hozac hmm, what's the point of exec in ( ) (vesync)?
1162563468 M * daniel_hozac or did you mean { and }?
1162563483 M * Hollow otherwise exec would replace the current process, but we want the overlay sync to take place too ..
1162563528 M * Hollow we could make it mutual exclusive of course ..
1162563544 M * daniel_hozac so what's the point of exec then? ;)
1162563563 M * daniel_hozac might as well just do && \ $_VSOMETHING....
1162563576 M * Hollow what's the point of doing exec there anyway in the other scripts?
1162563589 M * daniel_hozac to forward the return code, i think.
1162563617 M * Hollow well, bash automatically returns the code rom the last statement
1162563634 M * daniel_hozac and having the script around isn't really necessary after it's executed vsomething, so it might as well be replaced.
1162563652 M * Hollow yeah, that would make sense at least ..
1162563655 M * matti daniel_hozac, Hollow :)
1162563658 A * matti waves...
1162563664 A * Hollow waves back
1162563668 M * daniel_hozac hello matti
1162563732 M * Hollow daniel_hozac: ok, i removed exec from the emerge call, and removed the subshell from rsync, so rsync will replace it at least
1162563749 M * daniel_hozac didn't remove the subshell from emerge?
1162563763 M * Hollow i did :)
1162563828 M * daniel_hozac ok, looks fine to me then.
1162563831 M * Hollow maybe we should save the return code and return it after rsync? so in case emerge failed but rsync succeeded we still return failure?
1162563862 M * daniel_hozac yeah, i was thinking that too.
1162563867 M * Hollow but wouldn't work with exec then..
1162563874 M * daniel_hozac nope.
1162563884 M * Hollow but probably better anyway
1162563889 M * daniel_hozac yeah, i'd say so.
1162564065 M * Hollow ok, reload the patch
1162564100 M * Hollow ar..
1162564104 M * Hollow s/-o/-a/
1162564334 M * daniel_hozac yeah, looks fine.
1162564347 M * Hollow ok, i'll commit it then
1162564359 M * daniel_hozac or, hmm.
1162564393 M * daniel_hozac nah, brain misfire, it's fine.
1162564402 M * Hollow ok :)
1162564503 M * matti :)
1162564528 M * Hollow only thing left: adapt init script to vsyswrapper
1162564577 M * daniel_hozac hmm?
1162564590 M * Hollow the gentoo init script
1162564594 M * daniel_hozac vserver-wrapper, i guess?
1162564599 M * Hollow yeah ..
1162564608 M * daniel_hozac vsysvwrapper is the legacy "run host services in a chbind" script ;)
1162564618 M * Hollow brain misfire ;)
1162564621 M * daniel_hozac hehe.
1162564799 M * Hollow oh, i have already done it
1162564800 M * Hollow awesome
1162564801 M * Hollow :D
1162564812 M * daniel_hozac lol
1162565399 J * kir_home ~kir@swsoft-mipt-nat.sw.ru
1162566158 M * daniel_hozac i added the files to contrib/manifest.dat.pathsubst as well.
1162566265 M * phreak`` hrm, no 2.1.1 ? :|
1162566271 M * daniel_hozac not yet.
1162566274 A * phreak`` runs from Bertl_zZ's revenge
1162566283 M * phreak`` daniel_hozac: ;)
1162566296 M * daniel_hozac -rc47 should be out RSN, and then hopefully released as 2.1.1 unmodified ;)
1162566328 A * phreak`` ain't stupid even if it might appear to be sometimes :P
1162566367 M * daniel_hozac hehe.
1162566421 M * phreak`` bah, 25 EUR for winter tires ..
1162566713 J * gerrit ~gerrit@c-67-160-146-170.hsd1.or.comcast.net
1162567532 Q * renihs Quit: Leaving
1162567818 J * ntrs ~ntrs@68-188-55-120.dhcp.stls.mo.charter.com
1162567850 J * FireEgl FireEgl@Sebastian.Atlantica.US
1162568436 J * ntrs_ ~ntrs@d3-90.rb.ot.centurytel.net
1162568451 M * Hollow daniel_hozac: ah yeah.. the manifest..
1162568469 M * Hollow currently writing a big blog entry ;)
1162568524 M * daniel_hozac hehe, cool
1162568534 M * daniel_hozac i always forget the manifest too....
1162568779 M * wam when I give a vserver LOTS of VM (rlimits/as), say 1-2 GB, will this really take memory or is this just the reservable limit for virtual memory?
1162568836 Q * ntrs Ping timeout: 480 seconds
1162569176 Q * kir_home Read error: Operation timed out
1162569319 M * Hollow daniel_hozac: uhm ...
1162569321 M * Hollow zeus vservers # vserver gentoo3 build --context 44 --initstyle gentoo -m template -- -d gentoo -t /var/vservers/stage3-i686-20060317.tar.bz2
1162569321 M * Hollow >>> Adding shared /usr/portage to fstab ...
1162569321 M * Hollow >>> Checking init-style ... gentoo
1162569321 M * Hollow Unsupported packaging method: application/empty (application/x-bzip2)
1162569368 M * daniel_hozac can you extract it with tar -xjf?
1162569389 M * Hollow yeah ..
1162569391 M * daniel_hozac wam: nothing is ever reserved, it's a limit, not a guarantee.
1162569404 M * daniel_hozac Hollow: hmm, strange that file would return that then...
1162569420 M * Hollow zeus vservers # file stage3-i686-20060317.tar.bz2
1162569420 M * Hollow stage3-i686-20060317.tar.bz2: bzip2 compressed data, block size = 900k
1162569426 M * daniel_hozac file -z
1162569448 M * Hollow ah yeah.. empty .. wtf
1162569475 M * daniel_hozac that's very strange.
1162569559 J * shedi ~siggi@inferno.lhi.is
1162570012 Q * FaUl Ping timeout: 480 seconds
1162570589 N * Bertl_zZ Bertl
1162570593 M * Bertl morning folks!
1162570623 P * wam Verlassend
1162570639 M * Hollow morning Bertl
1162570786 J * stefani ~stefani@208.99.202.100
1162571094 M * Bertl hola stefani!
1162571226 M * stefani salut.    Buna dmineatsa
1162571428 Q * ruskie Remote host closed the connection
1162571838 J * ruskie ~ruskie@ruskie.user.oftc.net
1162572015 M * Bertl stefani: multumesc! (took me some time to figure the language and translate that :)
1162572080 M * stefani todah rabah !
1162572174 M * Bertl do you speak all those languages or are you collecting phrases?
1162572307 M * stefani some.
1162572356 M * stefani some of both, really.  learned arabic, french, farsi, hebrew, and a bit of turkish.   phrases from spanish, portugese, romanian, et al .
1162572391 M * Bertl arabic, nice .. they write from right to left, yes?
1162572407 J * oli ice@p54A3B7AD.dip0.t-ipconnect.de
1162572409 M * oli hi all
1162572413 N * oli crypt
1162572416 M * daniel_hozac hello
1162572417 M * Bertl welcome crypt!
1162572425 M * crypt thx :)
1162572435 M * crypt derjohn, are you online?
1162572856 M * derjohn crypt, looks like ;)
1162572960 M * crypt nice
1162573234 M * Hollow http://planet.gentoo.org/developers/hollow/2006/11/03/baselayout_vserver_is_dead
1162573386 M * daniel_hozac Hollow: awesome!
1162573391 J * immo ~immo@miniFaUl.wlan.chaostreff-dortmund.de
1162573392 M * Hollow indeed :)
1162573393 M * immo weha
1162573396 N * immo FAUl
1162573399 N * FAUl FaUl
1162573410 M * FaUl anyone with kernel-debugging knowledge around? bertl
1162573410 M * FaUl ?
1162573448 M * FaUl i just got a crashed kernel (unfortunately without any serial console attached while crashing) and got a lot of stuff on the monitor
1162573525 M * daniel_hozac camera?
1162573528 M * FaUl i wonder what i have to transcript to make it possible that someone "who knows"[tm] can debug that
1162573531 M * FaUl nope, none here
1162573545 M * FaUl it says lot of stuff like (#d478,*1):c0117810 clr_vx_info f2de6000[#9,85.32] @de9131c0
1162573555 M * daniel_hozac vserver history tracing.
1162573578 M * daniel_hozac do you see the first few lines? like OOPS/BUG/panic?
1162573602 M * FaUl and last line is some position of EIP: [<c0404c58>] __lock_text_start+0x0/0xf SS:ESP 0068:c3ff5e10
1162573620 M * FaUl nope, that one i pasted is the first one
1162573632 M * FaUl i may check the log, it is on another mashine luckily
1162573652 M * daniel_hozac please do.
1162573700 M * FaUl aii
1162573726 M * FaUl someone forgot to start the syslogd on the other host *NARF*
1162573778 M * FaUl fuck
1162573795 M * FaUl is there any way to recover that stuff that is outside the display?
1162573809 M * daniel_hozac shift page up doesn't work, i assume?
1162573831 M * FaUl nope
1162573836 M * FaUl unfortunately not
1162573861 M * daniel_hozac it usually doesn't. there's no way that i'm aware of.
1162573878 M * FaUl hm
1162573903 M * Bertl do you have the built kernel?
1162573908 M * FaUl is there any chance to hunt down that bug without the oops/panic?
1162573916 M * FaUl Bertl: yes, it's on that disk, i just have to restart
1162573922 M * Bertl if so, try addr2line -e vmlinux c0404c58
1162573931 M * FaUl and i thought ask first if that history is importent ;-)
1162573945 M * Bertl yeah, sure the history is
1162573956 M * FaUl so i have to transcript it?
1162573964 M * Bertl if you can transcribe that to paste.linux-vserver.org, that'd be great!
1162573973 M * FaUl ok, so i'll do
1162574089 M * matti Hello Bertl :)
1162574098 M * Bertl hey matti!
1162574116 M * matti :)
1162574121 M * FaUl ok, I hope that i'm not going into some transcription-failures ;-)
1162574462 M * FaUl how is that history structurized btw?
1162574468 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-rc47, 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 ;)
1162574478 M * Bertl *rc47 to be released as 2.1.1
1162574482 M * matti :>
1162574485 M * FaUl wheeahaa
1162574486 M * mnemoc :D
1162574494 A * mnemoc does the dance of joy
1162574548 M * derjohn aaaaaaaaaaaaaah
1162574558 M * derjohn just finished rc46 packages !
1162574567 M * matti Hehh
1162574569 M * derjohn well, will build again !
1162574572 M * matti :>
1162574573 M * FaUl Bertl: what are the various fields for? (#d436,*0):c01175bc set_vx_info f2de6000[#9,86.33[ @f42c5e00
1162574580 M * derjohn Bertl, changes ?
1162574588 M * derjohn crypt, read !
1162574780 M * Bertl derjohn: http://paste.linux-vserver.org/617 (vs2.1.0-vs2.1.1)
1162574805 M * Bertl derjohn: compared to rc46, debug.txt, base duplicate inclusion warning removed, minor whitespace cleanups
1162574826 M * daniel_hozac Bertl: fs/super.c printk, that's not intentional, is it?
1162574886 M * derjohn Bertl, wasnt there some CoW stuff you and daniel_hozac talked 'bout ? (chown or such)
1162574898 M * daniel_hozac that's been merged for a few rcs now.
1162574904 M * Bertl daniel_hozac: ops, no, that sneaked in
1162574904 M * daniel_hozac and we've been fixing problems with it ;)
1162574942 M * Bertl okay, removed the mount printk
1162574974 M * Bertl and updated in place :)
1162575004 M * Bertl http://plm.testing.osdl.org/patches/show/Linux-VServer-2.6.18.1-vs2.1.1-rc47
1162575048 M * Bertl okay, translocating now .. will be back in a few hours
1162575052 M * Bertl have fun! cya!
1162575057 N * Bertl Bertl_oO
1162575538 M * FaUl finally: http://paste.linux-vserver.org/618
1162575542 M * FaUl should be accurate
1162575585 M * FaUl ok, i'll reboot now
1162575971 M * FaUl what does "vxW: xid=11 messing with the procfs" mean?
1162576243 M * daniel_hozac xid 11 tried to change something in proc.
1162576249 M * daniel_hozac permissions, contents, ownership, etc.
1162576554 M * FaUl weltzentrale:/usr/src/linux-2.6.17.13# addr2line -e vmlinux c0404c58
1162576555 M * FaUl include/asm/spinlock.h:61
1162576555 M * FaUl weltzentrale:/usr/src/linux-2.6.17.13# 
1162576560 M * FaUl thats what bertl requested
1162576578 M * FaUl ah, its weltzentrale:/usr/src# uname -a
1162576579 M * FaUl Linux weltzentrale 2.6.17.13-vs2.1.1-rc35 #31337 SMP Fri Sep 29 20:36:05 CEST 2006 i686 GNU/Linux
1162576582 M * FaUl btw
1162576676 M * daniel_hozac well, without some more information (like what tried to take the spinlock), it's gonna be hard to debug that.
1162576716 M * FaUl yea, guess so
1162576717 M * FaUl ;-(
1162576823 A * FaUl will try to get some serial-console attached somehow
1162577181 M * FaUl what does that security-marking stuff do?
1162577247 M * FaUl ah, ic, its not vserver but linux virtual server
1162577271 P * marcfiu 
1162577466 Q * bronson Ping timeout: 480 seconds
1162578743 Q * derjohn2 Ping timeout: 480 seconds
1162578744 J * derjohn2 ~aj@dslb-084-059-024-044.pools.arcor-ip.net
1162578805 M * daniel_hozac Bertl_oO: vcmd_bcaps, shouldn't that be vcmd_bcaps_v0?
1162578842 Q * FireEgl Quit: Bye...
1162579392 M * FaUl ok, how likely is it that the syslogd can export the dump over network to another syslogd in case of a panic 
1162579455 M * daniel_hozac netconsole is probably better for that.
1162580328 M * FaUl mhh, i'll try to figure something out
1162580660 Q * FaUl Quit: Lost terminal
1162581621 J * bronson ~bronson@66.160.177.200
1162582186 Q * gerrit Ping timeout: 480 seconds
1162583130 J * nyu ~rmh@87.219.167.193
1162583135 M * nyu hi
1162583150 M * nyu does /etc/vservers/net2/interfaces/dev refer to the host or the guest interface name?
1162583167 M * nyu and, does /etc/vservers/net2/interfaces/0 mean eth0?
1162583184 M * micah nyu: no, and no
1162583191 M * nyu is there documentation about that somewhere?  I couldn't find it
1162583211 M * micah nyu: it depends on what you put in there, the 'dev' file will contain a device on the host
1162583245 M * micah the 0 is just an arbitrary number that you can assign, if you put a dev file inside 0/ with 'eth0' in it, then it will be that device
1162583252 M * micah nyu: have you looked at the flower page?
1162583253 M * nyu I want to give the guest full control of this interface (the host doesn't use it at all)
1162583309 M * nyu what flower page?
1162583394 M * nyu oh, right.  it needed a restart
1162583403 M * daniel_hozac http://www.nongnu.org/util-vserver/doc/conf/configuration.html
1162583435 M * daniel_hozac there's no such thing as a "guest interface name".
1162583451 M * daniel_hozac the interface name will be the same as that of the host that carries the IP address(es) your guest has access to.
1162583461 M * nyu ah, ok
1162583475 M * nyu is the host always in charge for assigning IP addresses to it?
1162583482 M * nyu I'd like to run dhclient from the guest
1162583494 M * daniel_hozac networking happens on the host. guests are just limited to a subset of IP addresses.
1162583507 M * nyu arg, my eyes hurt ;)
1162583514 M * daniel_hozac change the stylesheet ;)
1162583529 M * daniel_hozac (i probably should change the default... we have quite a few nice stylesheets already)
1162583556 M * nyu uhm that's problematic
1162583580 M * nyu if I setup that interface via dhcp on the host, then the routing table is modified to use that interface for 0.0.0.0
1162583592 M * nyu I can revert/avoid that, but then it won't work for the guest
1162583600 M * nyu can guest and host have different routing tables?
1162583642 M * daniel_hozac they can use different routing tables, with source-based routing.
1162583729 M * nyu oh, source-based routing sounds interesting
1162583864 N * Bertl_oO Bertl
1162583870 M * Bertl back now ...
1162583905 M * nyu if I setup source-based routing, I guess then I can use chbind to assign a process to a route or another, without need for a full vserver guest?
1162583923 M * Bertl yep
1162583961 M * daniel_hozac FYI: util-vserver now has 2.1 support ;)
1162583976 M * nyu sounds good
1162583983 M * Bertl daniel_hozac: ah? details?
1162583991 M * daniel_hozac well, this is just the basics.
1162584024 M * daniel_hozac but vattribute --xid ... --bcap ... will now raise the available bcaps.
1162584046 M * daniel_hozac and the new scheduler should be configurable.
1162584064 M * daniel_hozac (configuration for that is next on the list...)
1162584096 M * Bertl excellent work! tx!
1162584150 J * gerrit ~gerrit@bi01p1.co.us.ibm.com
1162584171 Q * sladen Ping timeout: 480 seconds
1162584334 J * sladen paul@starsky.19inch.net
1162584455 M * Bertl wb sladen! welcome gerrit!
1162584978 M * Bertl okay, rescheduled rc47 (forgot to do that after the update)
1162585094 Q * nyu Remote host closed the connection
1162585302 M * sladen Bertl: groovy.
1162585303 Q * bronson Quit: Ex-Chat
1162585405 M * gerrit Hi Bertl!
1162585416 J * nyu ~rmh@87.219.167.193
1162585630 J * dna_ ~naucki@3-222-dsl.kielnet.net
1162585766 Q * nyu Quit: leaving
1162586032 Q * dna___ Read error: Operation timed out
1162586792 Q * eyck Remote host closed the connection
1162586794 J * eyck eyck@ghost.anime.pl
1162587044 J * Piet hiddenserv@tor.noreply.org
1162587218 M * daniel_hozac Bertl: by the presence of fill-rate2 and interval2, one can infer VXSM_IDLE_TIME, right?
1162587353 M * Bertl yes, I'd suggest to do so
1162587372 M * Bertl i.e. the flag should IMHO be set by the tools when one of them is present
1162587410 M * Bertl and I'd suggest to assume I2/R2 = I/R when onely one is specified 
1162587431 M * Bertl (or alternatively I=I2,R=R2
1162587448 M * Bertl okay, off for a shower now .. back shortly
1162587482 M * daniel_hozac so you have to set the I2 and R2 even though IDLE_TIME isn't set? 
1162587743 M * Hollow isn't idle time configured in the kernel? and you only tell the syscall to use i and r for idle time by setting vxsm_idle_time?
1162587774 M * daniel_hozac you also have to set VXSM_FILL_RATE2|VXSM_INTERVAL2.
1162587786 M * Hollow hm ..
1162587795 M * daniel_hozac (that's what sets the idle time variables)
1162587808 M * daniel_hozac and AFAICT, you have to set VXSM_IDLE_TIME to enable it.
1162587840 M * Hollow yeah, you're right ..
1162587941 J * bronson ~bronson@66.160.177.200
1162588325 J * Aiken ~james@tooax6-218.dialup.optusnet.com.au
1162589635 M * Bertl back now
1162589983 Q * Rich_Estill Quit: Leaving
1162590462 M * Bertl ah, welcome bronson! morning Aiken!
1162590620 M * Aiken hi
1162590653 M * bronson Hi Bertl.
1162590709 M * Bertl Aiken: everything working fine for you now? new dietlibc and so?
1162590819 M * Aiken it seems fine with the new dietlibc, the main problem had been vlogin which now works
1162590838 M * Bertl okay, any remaining issues you encountered/observed?
1162590909 M * Bertl daniel_hozac: will there be a 0.30.212 release accompanying the 2.1.1 release? or should I prepare 211 for that?
1162590919 M * Aiken have not tried rc46 or 47 yet
1162590940 M * Bertl would be nice to get feedback, but I doubt it really changes much for you
1162591043 M * Aiken I'll run up a rc47 this morning
1162591156 M * daniel_hozac Bertl: 212 isn't quite ready yet.
1162591216 Q * bronson Ping timeout: 480 seconds
1162591228 J * bronson ~bronson@adsl-64-161-106-11.dsl.snfc21.pacbell.net
1162591242 M * Bertl daniel_hozac: okay, 211 should be fine, I guess?
1162591252 M * daniel_hozac yep.
1162591259 M * Bertl daniel_hozac: or do you have any patches you'd like me to add there?
1162591499 M * daniel_hozac no really important changes, AFAICT.
1162591616 Q * gerrit Ping timeout: 480 seconds
1162592373 J * gerrit ~gerrit@bi01p1.co.us.ibm.com
1162592411 Q * sladen Ping timeout: 480 seconds
1162592572 J * dreamind apwdsl@p548AB06D.dip0.t-ipconnect.de
1162592585 M * dreamind Hi folks :D
1162592606 M * Bertl hey dreamind!
1162592656 J * sladen paul@193.28.45.41
1162594044 M * Aiken Bertl it boots, a guest starts and stops. No more xid mismatch messages
1162594066 M * Bertl great!
1162594277 J * node_ ~dwindsor@c-69-143-154-220.hsd1.md.comcast.net
1162594306 Q * bronson Ping timeout: 480 seconds
1162594916 Q * meandtheshell Quit: exit (0);
1162595559 Q * dna_ Quit: Verlassend
1162596328 J * Zaki ~Zaki@212.118.125.242
1162596336 M * Bertl wb Zaki!
1162597166 M * cehteh i have some processes hanging in a sync() (ok, it is a stupid idea to zero out a 80GB HD over USB 1.1) ... but even a kill -9 cant kill them, is that vserver specific or does that apply to vanilla linux too?
1162597266 M * Bertl that is quite normal
1162597290 M * Bertl it will come back once the 'buffers' are writen or the device gives an error
1162597433 J * bronson ~bronson@66.160.177.200
1162597443 M * Bertl wb bronson!
1162597455 M * bronson :)
1162597464 M * bronson Stupid unreliable wireless network...
1162597561 J * almak ~almak@192.83.249.36
1162597614 M * almak hi
1162597630 M * Bertl hey almak!
1162597636 M * almak hey
1162597646 M * almak question about bandwidth limit. 
1162597684 M * almak I hear that it is possible to use htb to rate limit miltiple Vserver contexts. Question I have is how does the match syntax work?
1162597704 M * Bertl the guests are bound to certain ips
1162597721 M * Bertl so, you basically select to/from ips accrodingly
1162597744 M * Bertl all traffic local to the host will use lo, so make sure to exclude that one
1162597773 M * almak I see. I was under the impression that I can match based on context id.
1162597816 M * cehteh Bertl: ok .. well .. hangs since hours ;)
1162597824 M * Bertl almak: in theory you could, but there is no real benefit over the ips (and it requires patching iptables and tools)
1162597928 M * almak So if vserver contexts share the same IP then the filter cannot distinguish.
1162597973 M * Bertl depends, outgoing traffic would be fine, incoming could not be classified correctly (well, only too late, when delivered to a socket)
1162598297 M * almak still don't get it. If I have a single IP how does the filtering work?
1162598330 M * Bertl well, if you _have_ a single ip, you are better off if you use S/DNAT, otherwise services will clash
1162598337 M * Bertl let me give an example here:
1162598358 M * Bertl let's assume you have a single public ip, and 3 guests
1162598389 M * Bertl if you put apache on all of them, they will have a problem, unless you use different ports for each of them
1162598393 M * Bertl right?