1174176037 M * daniel_hozac ah right.
1174176048 M * doener phew :)
1174176254 N * DoberMann DoberMann[ZZZzzz]
1174176273 M * Bertl DoberMann[ZZZzzz]: good night!
1174176355 Q * bonbons Quit: Leaving
1174176373 M * daniel_hozac doesn't seem like the parent process is getting SIGHUP'ed at all though.
1174176493 Q * meandtheshell Quit: Leaving.
1174177013 M * doener daniel_hozac: hm, cross context signal barrier?
1174177033 M * daniel_hozac IIRC we checked for that.
1174177036 M * doener ok
1174177055 M * doener it's from 0 -> XXX anyway
1174177061 M * daniel_hozac yeah.
1174177165 M * doener hm, my "login" process just stays in select
1174177178 M * daniel_hozac is your sshd process really gone?
1174177190 M * doener just the "su"
1174177197 M * daniel_hozac (i found that i have to kill that as well)
1174177204 M * daniel_hozac sometimes.
1174177207 M * daniel_hozac sometimes i don't.
1174177217 M * doener oh wow, a bazillion of  "vlogin: tcsetattr(): I/O error" on ssh logout
1174177234 M * daniel_hozac heh, i just got those for the first time ever.
1174177307 M * doener and this time, the ssh logout just hangs and "login" is stuck in select() again
1174177363 M * daniel_hozac yeah, it seems to be really racy.
1174177373 M * daniel_hozac and all sorts of different issues.
1174177378 M * doener ah, forcibly killing ssh got my the loop
1174177397 M * daniel_hozac sometimes i get it into a state where every keypress is executed separately.
1174177425 M * daniel_hozac (and seemingly they two shells take one character each)
1174177431 M * doener sounds like a missing terminal_reset()
1174177461 M * daniel_hozac right, but how would that happen?
1174177469 M * doener ah, jumping between processes via "vserver xxx enter" haven't seen that since .209 or so
1174177488 M * daniel_hozac it's in atexit and terminal_end .
1174177522 M * doener back then, su - followed by "vserver xxx enter" and some other condition, made me jump between the su - shell and my ssh login shell, totally nuts
1174177522 M * daniel_hozac TBH i don't even see how it's possible with vlogin.
1174177530 M * daniel_hozac yeah.
1174177537 M * daniel_hozac exactly that's happening.
1174177590 M * doener that would be an older, non-vlogin, problem then though
1174177610 M * doener I actually thought that that was fixed by vlogin :)
1174177622 J * s0undt3ch ~s0undt3ch@80.69.34.154
1174177627 M * daniel_hozac i'd expect it to be.
1174177658 M * doener maybe the unclean death makes the problem return as a sideeffect
1174177680 M * daniel_hozac but even so, vlogin is still in there.
1174177703 M * daniel_hozac that should restrict the guest's capabilities to mess with the host shell.
1174178973 M * doener daniel_hozac: hm, SIGHUP is sent to the session leader which is "bash", which will forward it to _its_ children
1174178988 M * doener but the child process was "su -", not "login"
1174178995 M * doener so the signal never arrives
1174179004 M * doener well, it's never even sent to it
1174179007 M * doener sounds logical?
1174179015 M * daniel_hozac hmm, yeah.
1174179492 M * doener sending a SIGHUP to either "login" or "bash -login" works as expected
1174179550 M * doener sending it to "su -" results in "Session terminated, killing shell..." and a messed up terminal
1174179587 M * doener and "login" continues to run
1174179679 M * doener and when "-su" is signalled, the terminal is also messed up, but "login" dies off
1174179748 M * doener http://www.mail-archive.com/bug-bash@gnu.org/msg02585.html
1174179906 M * doener and indeed, that only happens when "su" comes into play, just starting a second bash is not enough
1174180032 M * doener does anyone know where to look for their VCS? We can probably just look up how they fixed that
1174180055 M * doener I feel even more lost on gnu.org than on mozilla.org...
1174180089 M * daniel_hozac savannah.gnu.org?
1174180112 M * doener has no files in the bash repo (or I'm too dumb to find them)
1174180123 M * daniel_hozac true...
1174180128 M * daniel_hozac no idea then.
1174180283 M * doener well, then again, the bug is said to be in readline anyway...
1174180307 M * doener which has 0 files as well... *sigh*
1174180313 M * daniel_hozac hehe
1174180358 M * Bertl doener: experimenting with alternative repo systems?
1174180370 M * doener me?
1174180377 M * daniel_hozac well, i'm quite tired so i'm gonna get some sleep. if/when ;) you figure out how to fix it, please let me know. good night!
1174180410 M * doener daniel_hozac: actually, I'm pretty sure the (ret = read()) == 0 fix is fine
1174180454 M * doener I'd just like to see the readline fix as a confirmation
1174180527 M * doener that said, I'm off to bed as well
1174180533 M * doener cya
1174180581 M * Bertl good night folks! I'm pretty much off to bed too
1174181530 J * brcc_ bruce@i.am.someasshole.com
1174182209 M * Bertl okay, off to bed now .. have a good one everyone!
1174182213 N * Bertl Bertl_zZ
1174183236 Q * hallyn Ping timeout: 480 seconds
1174184464 Q * brcc_ Ping timeout: 480 seconds
1174184494 Q * duckx Remote host closed the connection
1174185484 Q * softi42 Ping timeout: 480 seconds
1174185883 J * FireEgl Proteus@adsl-61-136-122.bhm.bellsouth.net
1174186512 J * softi42 ~softi@p549D593D.dip.t-dialin.net
1174186536 Q * zLinux Ping timeout: 480 seconds
1174186559 J * zLinux ~zLinux@88.213.58.16
1174189827 J * infowolfe ~infowolfe@c-67-164-195-129.hsd1.ut.comcast.net
1174190801 Q * ensc Killed (NickServ (GHOST command used by ensc_))
1174190811 J * ensc ~irc-ensc@p54B4D0C9.dip.t-dialin.net
1174191182 J * brcc_ bruce@i.am.someasshole.com
1174193344 J * LOLERSKATES qweeasd@cpe-075-183-162-181.sc.res.rr.com
1174193346 M * LOLERSKATES DCC SEND 0123456789123456
1174193348 P * LOLERSKATES 
1174193417 J * LOLERSKATES qweeasd@cpe-075-183-162-181.sc.res.rr.com
1174193418 M * LOLERSKATES DCC SEND 420SMOKEWEED420
1174193420 P * LOLERSKATES 
1174194286 N * DoberMann[ZZZzzz] DoberMann
1174195099 J * DoberMann_ ~james@AToulouse-156-1-132-60.w90-30.abo.wanadoo.fr
1174195208 Q * DoberMann Ping timeout: 480 seconds
1174195490 J * jamesdell ~net@bb121-6-37-6.singnet.com.sg
1174195919 P * jamesdell 
1174197397 N * DoberMann_ DoberMann[PullA]
1174198465 Q * zLinux osmotic.oftc.net oxygen.oftc.net
1174198465 Q * s0undt3ch osmotic.oftc.net oxygen.oftc.net
1174198465 Q * michal` osmotic.oftc.net oxygen.oftc.net
1174198465 Q * cdrx osmotic.oftc.net oxygen.oftc.net
1174198465 Q * derjohn osmotic.oftc.net oxygen.oftc.net
1174198465 Q * CHTEKK osmotic.oftc.net oxygen.oftc.net
1174198465 Q * micah osmotic.oftc.net oxygen.oftc.net
1174198473 J * zLinux ~zLinux@88.213.58.16
1174198473 J * s0undt3ch ~s0undt3ch@80.69.34.154
1174198473 J * michal` ~michal@www.rsbac.org
1174198473 J * CHTEKK ~chtekk@84.55.211.45
1174198473 J * derjohn ~derjohn@80.69.41.3
1174198473 J * cdrx ~legoater@cap31-3-82-227-199-249.fbx.proxad.net
1174198473 J * micah ~micah@micah.riseup.net
1174202812 Q * dghill Ping timeout: 480 seconds
1174205319 J * herka ~herka@8.111.101-84.rev.gaoland.net
1174205381 Q * cdrx osmotic.oftc.net oxygen.oftc.net
1174205381 Q * derjohn osmotic.oftc.net oxygen.oftc.net
1174205381 Q * CHTEKK osmotic.oftc.net oxygen.oftc.net
1174205381 Q * s0undt3ch osmotic.oftc.net oxygen.oftc.net
1174205381 Q * zLinux osmotic.oftc.net oxygen.oftc.net
1174205381 Q * michal` osmotic.oftc.net oxygen.oftc.net
1174205381 Q * micah osmotic.oftc.net oxygen.oftc.net
1174205470 J * zLinux ~zLinux@88.213.58.16
1174205470 J * s0undt3ch ~s0undt3ch@80.69.34.154
1174205470 J * michal` ~michal@www.rsbac.org
1174205470 J * CHTEKK ~chtekk@84.55.211.45
1174205470 J * derjohn ~derjohn@80.69.41.3
1174205470 J * cdrx ~legoater@cap31-3-82-227-199-249.fbx.proxad.net
1174205470 J * micah ~micah@micah.riseup.net
1174205819 Q * herka Quit: herka
1174205829 J * herka ~herka@8.111.101-84.rev.gaoland.net
1174205869 Q * herka 
1174206695 J * Piligrim ~mva@234-216-124-91.pool.ukrtel.net
1174206947 Q * cdrx Quit: Leaving
1174207649 Q * Aiken Read error: Connection reset by peer
1174208133 Q * DreamerC Ping timeout: 480 seconds
1174208150 J * dna ~naucki@p54BCFB2A.dip.t-dialin.net
1174209277 J * meandtheshell ~markus@85-124-232-240.work.xdsl-line.inode.at
1174209619 J * bonbons ~bonbons@83.222.39.9
1174215117 Q * Piligrim Quit: õÈÏÖÕ
1174216448 J * duckx ~Duck@tox.dyndns.org
1174216736 M * matti :)
1174217659 J * DavidS ~david@chello062178045213.16.11.tuwien.teleweb.at
1174217663 J * boci^ boci@pool-7377.adsl.interware.hu
1174218633 J * andres ~as.sa@rtfm.ping.ee
1174218779 M * Hollow matti: :)
1174218822 M * matti Hi Hollow :)
1174218825 M * matti How are you?
1174218842 M * Hollow fine, finally back in berlin and a working machine :)
1174218845 M * Hollow (now amd64 ;)
1174218855 M * matti Whooo :)
1174218870 M * matti I can help testing stuff on my Core 2 Duo machine.
1174218880 M * matti If somebody need such thing.
1174218937 M * Hollow hehe, it rocks.. i also bought a 22-inch widescreen tft :)
1174218961 A * DavidS drools
1174218973 M * matti And I am trying to buy one of my companies old 1U machines for us :) Dual Xeon 3 GHz, 4 GB of RAM. But, this may take w while... Too much bureaucracy.
1174218976 M * matti Hollow: LOOL
1174218992 M * matti Hollow: You're crazy mate :)
1174219001 M * Hollow matti: 1680x1050 rocks :P
1174219011 M * matti -_-'
1174219033 M * matti My 15.4" lnotebook can produce 1900x1200
1174219048 M * Hollow especially with the old 19-inch tft on its left it :P
1174219054 M * Hollow hehe
1174219054 M * matti But, it is hard to read letters then.
1174219057 M * phreak`` Hollow: for that resolution you need a 22" ? :P
1174219066 A * phreak`` pats his 15,4" notebook
1174219068 M * matti phreak``: No... He is just crazy.
1174219076 M * matti phreak``: :)
1174219079 M * phreak`` matti: true point, as he is crazy *g*
1174219085 M * matti :>
1174219092 A * Hollow nods
1174219101 M * matti phreak``: Well, Bertl is an alien and Hollow is crazy.
1174219108 M * phreak`` *ugh*
1174219115 M * phreak`` since when ?
1174219120 M * matti I am glad, that people that use Vserver are not aware of that ;)
1174219129 M * phreak`` did Bertl get replaced by an alien ? :S
1174219135 M * matti Well.
1174219145 M * matti Elvis leaves Earth and go back to home.
1174219150 M * matti And then Bertl comes.
1174219150 M * matti :)
1174219185 M * matti You know the Roswell stuff? :)
1174219186 M * matti ;p
1174219202 M * phreak`` matti: mate, you are crazy too :P
1174219212 M * matti [ Sorry I will behave now ;ppp ]
1174219218 M * phreak`` heh :D
1174219219 M * matti phreak``: Yes! Yes in fact I am!
1174219229 M * matti phreak``: Don't say this loudly ;p
1174219248 M * matti phreak``: THEY watch us.
1174219255 M * matti phreak``: :D
1174219260 M * phreak`` matti: heh
1174219271 M * matti ;)
1174219286 M * phreak`` Hollow: hrm, any other *good* terminal for kde (besides konsole and yakuake) ?
1174219291 M * matti Just tell me if I am too annoying with my sick sense of humour :<
1174219307 M * phreak`` matti: when did somebody say something like that ?
1174219340 M * matti phreak``: Nobody... yet :)
1174219388 M * phreak`` <-- lunch
1174219409 M * matti phreak``: Enjoy :)
1174219480 M * Hollow phreak``: no.. but can you imagine yakuake at 100%x100% using 9pt clean bitmap font on the 22-inch? O.o
1174219482 M * Hollow :))
1174219509 M * matti Hahahah
1174219535 M * Hollow most of the compiler output even fits in one line now :D
1174219543 M * daniel_hozac lol
1174219546 M * matti ROTFL
1174219557 M * matti Hollow: :-)))))))))
1174219574 M * matti Hollow: That is a very good reason to buy such screen ;p
1174219594 M * Hollow indeed, i can imagine 3-4 vim :vsplits fit fine as well :)
1174219596 M * matti Hollow: Now I understand you completely :)
1174219648 M * daniel_hozac doener: i think something like that is fine as well.
1174220835 M * matti :]
1174220896 M * phreak`` Hollow: hrm, which package has the clean font ? I've been searching for that some time now :P
1174220946 J * DreamerC ~dreamerc@125-225-101-77.dynamic.hinet.net
1174220948 M * Hollow phreak``: media-fonts/font-schumacher-misc-1.0.0
1174222607 J * ema ~ema@rtfm.galliera.it
1174223355 J * HiLander ~losom@203.146.63.184
1174223751 J * lilalinux ~plasma@dslb-084-058-195-009.pools.arcor-ip.net
1174223844 Q * lilalinux Remote host closed the connection
1174223895 J * lilalinux ~plasma@dslb-084-058-195-009.pools.arcor-ip.net
1174224340 Q * HiLander 
1174225628 Q * yarihm Quit: This computer has gone to sleep
1174226366 N * Bertl_zZ Bertl
1174226375 M * Bertl morning folks!
1174226458 M * daniel_hozac morning Bertl!
1174226471 A * Bertl sends an ufo squad to bring in matti ...
1174226579 Q * infowolfe Ping timeout: 480 seconds
1174226658 M * Bertl daniel_hozac: how's going?
1174226707 M * daniel_hozac quite good, i think... you?
1174226757 M * Bertl probably fine, we'll see after breakfast ... :)
1174226943 M * bXi maybe the wrong place to ask this but what are the prefered HDD makes around here?
1174226960 M * Bertl daniel_hozac: we had a problem with sched_prio not working without sched_hard enabled or so ... do you happen to remember the details?
1174226974 M * Bertl bXi: I prefer working and fast ones :)
1174226991 M * bXi yeah but which ones are working :p
1174226998 M * bXi i know for a fact that maxtors die to easily
1174227010 M * Bertl it really really depends ...
1174227023 M * Bertl let me give a few examples
1174227031 M * bXi if i'm being a cheapskate i'd go for samsung right now
1174227102 M * Bertl I have a bunch of Maxtor 4G*J* sitting in the basement for, hmm, 5 years or so? not a single bad block or whatever
1174227110 M * bXi hmmm
1174227113 M * bXi how big are those?
1174227120 M * Bertl ~150G
1174227124 M * bXi hmm
1174227127 M * Bertl but
1174227132 J * infowolfe ~infowolfe@c-67-164-195-129.hsd1.ut.comcast.net
1174227142 M * bXi at the office we have a bunch of dead 40gb maxtors
1174227148 M * bXi 20 of them or so
1174227158 M * Bertl I had quite a number of replacement for similar drives in a data center
1174227196 M * bXi and my 160gb maxtor died after 9 months of so after being completly new (straight from the factory)
1174227197 M * Bertl my impression was, that the temperature and the quality of the power supply is very important
1174227225 M * bXi but it was RMA'd once already so i had some bad luck
1174227251 M * Bertl a friend of mine is running Maxtor 6Y160M0's and he has a new bad block every day
1174227261 M * bXi hehe
1174227270 M * bXi i head good things about seagate
1174227272 M * Bertl he is running them at 55°C is a not so well ventilated server
1174227278 M * bXi ouch
1174227288 M * Bertl so definitely there is a relation
1174227292 M * bXi i'm fixxing my girlfriends computer atm
1174227306 M * bXi and i got those hard disks down to 24C
1174227327 M * Bertl I'm currently very happy about the seagate 7000.10 series
1174227362 M * Bertl but the first bunch of those perpendicular sata II drives went straight to RMA
1174227391 M * bXi i should just go crazy and order myself 2x250gb IDE from seagate and 2x500GB SATA :p
1174227441 M * Bertl western digital OTOH, didn show up on my radar since the older scsi days, where the disks literally burned through the chassis
1174227490 M * Bertl but that is only my personal impression :)
1174227502 M * bXi i got 2 WD's here atm
1174227509 M * bXi (in gf's pc)
1174227794 M * harry Bertl: does it make sense to upgrade my 2.2 patch to rc18 ?
1174227804 M * harry or do i just wait for the release now
1174227830 M * Bertl it always makes sense to update
1174227841 M * Bertl and you save the time when you do the release, no?
1174227845 M * harry yeah, but if there is a release in 2 days from now
1174227861 M * harry it won't be used by anyone
1174227873 M * Bertl sure?
1174227881 M * harry btw. can you do the release for 2.6.19 too? ;)
1174227893 M * harry (any 2.6.19 up till then off course
1174227906 M * Bertl yes, release will be for both
1174227906 M * harry (there is no grsec for 2.6.20 kernels)
1174227914 M * harry tnx
1174228354 J * hallyn ~xa@adsl-75-2-80-15.dsl.chcgil.sbcglobal.net
1174228367 M * Bertl wb hallyn!
1174228909 Q * hallyn Ping timeout: 480 seconds
1174229387 M * DavidS has anybody experience with one of the mergemem patches and vservers?
1174229462 M * Bertl not that I know of .. do they clash somewhere?
1174229524 M * DavidS no ... i was just wondering .. I'm currently setting up several VServers with automated config management ... each running a fat ruby daemon ...
1174229544 M * DavidS now I'm undecided whether I should add another gig ram or try a software approach ..
1174229553 Q * lilalinux Remote host closed the connection
1174229639 M * Bertl DavidS: well, the merge mem makes sense in a Linux-VServer environment, but somebody should test if it really brings a benefit, when the guests are already unified
1174229679 M * DavidS my guests are not unified currently .. I'm still on debian's 2.0 
1174229701 M * Bertl well, unification is probably your best memory gain without any changes
1174229723 M * Bertl btw, do you have a link to a recent mergemem version?
1174229759 M * DavidS nope, I was just brainstorming ... and knowing that unification will help, improves my patience significantly :)
1174229874 M * daniel_hozac doener, Bertl, Hollow: http://people.linux-vserver.org/~dhozac/p/uv/experimental/delta-vlogin-eof.diff
1174229885 M * Bertl DavidS: the interesting detail is, one of the mergemem guys (Phillip) was a student of mine ...
1174229951 M * Hollow daniel_hozac: does this fix the issue with processes hanging in select?
1174229976 M * daniel_hozac it fixes the endless loop of select() read()
1174229994 M * DavidS heh :)
1174229996 M * daniel_hozac for cases where select doesn't return, i don't think there's anything we can do.
1174230013 M * Hollow ok
1174230026 M * Bertl daniel_hozac: is (or can you do that?) a version installed in princeton? if so, it will get some testing today ...
1174230057 M * daniel_hozac i'll commit the fixes i have queued up and do an rc5.
1174230058 J * yarihm ~yarihm@105-227.1-85.cust.bluewin.ch
1174230068 M * daniel_hozac (which should hopefully be released as 0.30.213)
1174230074 Q * yarihm 
1174230075 M * Bertl daniel_hozac: okay, let me know as soon as it is available
1174230299 Q * Johnnie Ping timeout: 480 seconds
1174230505 M * daniel_hozac ugh, removing that file dep from the template build method is moving higher on my TODO list every time i try to use it...
1174231243 Q * matti Ping timeout: 480 seconds
1174232025 J * Johnnie ~jdlewis@jdlewis.org
1174232537 M * daniel_hozac http://people.linux-vserver.org/~dhozac/t/uv-testing/util-vserver-0.30.213-rc5.tar.bz2
1174232538 Q * Johnnie Read error: Connection reset by peer
1174232560 J * Johnnie ~jdlewis@jdlewis.org
1174232560 M * Bertl daniel_hozac: tx
1174233075 M * daniel_hozac i _think_ externalize should work now...
1174233102 M * Bertl okay, will test shortly, have a strange issue now
1174233107 M * daniel_hozac oh?
1174233114 M * Bertl postfix claims that port 25 is already in use
1174233126 M * Bertl but I cannot find anything, neither on the host nor guest
1174233135 M * daniel_hozac on any guest?
1174233153 M * daniel_hozac or could it be like that strange apache issue where it tried to bind the same port twice?
1174233172 M * Bertl I thought about that, and did an strace, which shows a single bind attempt
1174233244 M * Bertl binding with nc succeeds though ...
1174233254 M * Bertl ah, no, it doesn't ... strange!
1174233327 M * Bertl I guess I have to upgrade the kernel and reboot the machine
1174235798 Q * duckx Remote host closed the connection
1174236652 J * duckx ~Duck@tox.dyndns.org
1174236732 Q * Piet Remote host closed the connection
1174236739 M * daniel_hozac Bertl: we seem to have an initpid bug.
1174236766 M * daniel_hozac (this is on 2.6.20.1-vs2.2.0-rc16 though)
1174236790 M * daniel_hozac i.e. vps doesn't show the host's init, but the guest's init twice.
1174236808 M * Bertl nice
1174236846 M * Bertl please investigate, I can't see the init issues anymore
1174236873 M * daniel_hozac i'll try to reproduce it on 2.6.20.3-vs2.2.0-rc18.
1174236883 M * Bertl this is such a trivial principle, and it gets worse and worse with every change to procfs
1174236943 Q * zLinux Remote host closed the connection
1174237243 M * daniel_hozac fs/proc/base.c:pid_revalidate, shouldn't we goto out_put if task->pid != requested pid?
1174237324 M * daniel_hozac or, vx_map_pid(task->pid) != requested pid, at least.
1174237336 M * Bertl yep, that looks more like it
1174237398 M * daniel_hozac i'll make sure it's still present in the latest version and then try that.
1174237411 M * Bertl okay, thanks a lot!
1174237554 M * phreak`` harry: hrm, could you *please* put something like "IndexOptions +NameWidth=60 +SuppressDescription" into your .htaccess file ? :)
1174237926 M * harry ?
1174237978 M * daniel_hozac so that filenames are visible ;)
1174238008 M * phreak`` exactly :)
1174238022 M * phreak`` harry: you're a genius :)
1174238024 M * harry done
1174238053 M * daniel_hozac phreak``: have you had a chance to try 0.30.213-rcX?
1174238070 M * phreak`` daniel_hozac: erm, there is a -rc ? :)
1174238088 M * phreak`` daniel_hozac: I haven't been looked at util-vserver in a long while :)
1174238094 M * phreak`` daniel_hozac: but sure, I'll try it :)
1174238096 M * daniel_hozac http://people.linux-vserver.org/~dhozac/t/uv-testing/util-vserver-0.30.213-rc5.tar.bz2
1174238116 M * daniel_hozac i'd just like to have someone actually try the Gentoo initscripts before i release it :)
1174239001 M * phreak`` daniel_hozac: hrm, the bash_completion still hasn't made its way in there ? :)
1174239194 M * phreak`` daniel_hozac: aaaand, how are you installing the gentoo init-script ? via the install-distribution rule/target ?
1174239237 M * phreak`` daniel_hozac: found it :)
1174239286 M * daniel_hozac no, i haven't added the bash completion.
1174239308 M * phreak`` (the --with-initscripts is what i found)
1174239371 M * daniel_hozac it should autodetect.
1174239397 M * Bertl hmm, with rc5, the guest processes weren't killed on system shutdown ...
1174239407 M * Bertl anything I have to configure to do that?
1174239419 M * daniel_hozac hmm, no, that should happen automatically assuming the script is run...
1174239445 M * Bertl for all guests/xids or just for those marked to autostart?
1174239460 M * daniel_hozac all of them.
1174239470 M * Bertl okay, we'll see, now checking disks :(
1174239480 M * daniel_hozac okay :/
1174239626 M * daniel_hozac what distribution is that?
1174239679 M * phreak`` daniel_hozac: looks like they work :)
1174239695 M * daniel_hozac phreak``: okay, thanks a lot!
1174239722 M * phreak`` daniel_hozac: but that set_helper thing is completely unneccessary :)
1174239730 M * daniel_hozac how so?
1174239765 M * phreak`` daniel_hozac: because even util-vserver installs a symlink to get the right location for the vshelper
1174239796 M * daniel_hozac right, but that was added because we were talking about dropping the default of /sbin/vshelper...
1174239817 M * phreak`` ah, don't know about that :)
1174239915 M * Bertl I think the 'default'is fine, similar to hotplug
1174239919 M * daniel_hozac Bertl: initpid bug still present in rc18, booting the fix now...
1174239970 M * Bertl but IMHO it is better to change the setting (sysctl) instead of putting a symlink there
1174240000 M * Bertl note: IMHO this setting is something the post install script (package) should do
1174240004 M * phreak`` daniel_hozac: anything against this: http://rafb.net/p/aBMY7f96.html ?
1174240064 M * daniel_hozac no.
1174240379 M * daniel_hozac committed.
1174240565 M * Bertl hmm, the vserver scripts are all off by default
1174240582 M * Bertl that probably explains why nothing got run
1174240669 M * Bertl but probably that is something distro specific isntall has to care about
1174240691 M * Bertl postfix works now as expected
1174240710 M * daniel_hozac yeah.
1174241020 Q * michal` Ping timeout: 480 seconds
1174241060 J * hallyn ~xa@75.2.80.15
1174241595 J * michal` ~michal@www.rsbac.org
1174242091 M * yang Bertl: Any more ideas for the mips server?
1174242288 M * daniel_hozac Bertl: hmm, any pointers for how to get the requested pid?
1174242359 M * Bertl yang: didn't find any time recently ... but the next steps are compiling a mainline kernel with the debian config and trying this
1174242380 M * Bertl daniel_hozac: no, not really
1174242460 M * yang Bertl: well 2.6.18 from the debian booted well, you can get the config from the debian website and try to tune it to work with vserver patch ?
1174242475 M * yang I think the version was IP22-4000
1174242493 M * Bertl it will work out of the box with a Linux-VServer kernel
1174242503 M * Bertl and I think we should stay with mainline kernels for now
1174242524 M * Bertl but I have no idea where the config for the debian kernel is
1174242536 M * Bertl so if you can track that down, I can compile a kernel with modules
1174242546 M * yang Bertl: it must be available on http://packages.debian.org
1174242555 M * yang I will look for it now
1174242779 M * Bertl thanks!
1174242799 M * Bertl daniel_hozac: we get a duplicate pid, yes?
1174242812 M * yang http://ftp.fi.debian.org/debian/pool/main/l/linux-kernel-di-mips-2.6/kernel-image-2.6.18-4-r4k-ip22-di_0.13_mips.udeb  Maybe this would contain the config files ? I can still boot the mips machine and get the file out by ftp...
1174242820 M * daniel_hozac yes, the guest's init shows up once for it's actual pid, and once for pid 1.
1174242824 M * Bertl and it shows '1' or different pid values?
1174242834 M * daniel_hozac (but it has the same pid both times)
1174242857 M * Bertl show me how it looks like, please
1174242926 M * daniel_hozac http://paste.linux-vserver.org/1325
1174242995 M * Bertl ah, so ps inside sees it correctly?
1174243016 M * daniel_hozac yeah, but then /proc/1 in xid 1 appears to refer to the guest's init.
1174243042 M * Bertl okay, that just means that the pid=1 wasn't dropped
1174243048 M * daniel_hozac exactly.
1174243049 M * Bertl (i.e. needs the revalidate)
1174243066 M * Bertl what about two different guests, when they do ps?
1174243080 M * Bertl do they get 'wrong' init tasks shown?
1174243082 M * daniel_hozac haven't tried that yet, i assume fixing this would fix them all ;)
1174243105 M * Bertl I guess the 'fix' is already there, just doesn't hit for xid=1
1174243126 M * daniel_hozac oh? where?
1174243146 M * Bertl let's see if it works across contexts first
1174243167 M * daniel_hozac ah, vx_proc_task_visible, right.
1174243495 M * daniel_hozac different guests see their own inits.
1174243502 M * daniel_hozac so it's working as expected.
1174243564 Q * renihs Ping timeout: 480 seconds
1174243681 M * daniel_hozac so we add an if vx_info_map_pid(task->vx_info, task->pid) == 1 goto invisible to vx_proc_task_visible, or?
1174243812 M * daniel_hozac no, that wouldn't work...
1174243938 M * Bertl no, we check for pid=1 and xid=0/1
1174243953 M * Bertl that should cover this cornercase
1174243960 M * Bertl (in the revalidate)
1174243986 M * daniel_hozac hmm?
1174244693 M * Bertl sec, checking the code now ...
1174244953 J * Piet hiddenserv@tor.noreply.org
1174245134 M * Bertl we don't see the 'real' init in vps, correct?
1174246386 N * DoberMann[PullA] DoberMann
1174246470 M * hardwire its a shame you can't make tap interfaces in a vserver
1174246502 M * Bertl hardwire: why?
1174246529 M * hardwire well.. in general its a shame you can't add/remove interfaces
1174246534 M * hardwire but having tuntap support would be nice 
1174246538 M * Bertl hardwire: why?
1174246541 M * hardwire I use vservers for reasons you shouldn't use vservers.
1174246552 M * hardwire but ssh/openvpn/etc.. all use tap interfaces
1174246571 M * Bertl and they work quite fine, no?
1174246595 M * Bertl btw, most of them use tun interfaces, for a good reason
1174246607 M * hardwire sorry
1174246613 M * hardwire they use tun or tap interfaces
1174246614 M * hardwire depending
1174246618 M * hardwire vserver supports neither
1174246626 M * hardwire which its not really supposed to either.
1174246630 M * hardwire its just a shame. :)
1174246634 M * Bertl the creation is not allowed, using them is fine
1174246643 M * hardwire porque?
1174246653 M * Bertl and you can allow the creation by adding the capabilities too
1174246658 M * hardwire how?
1174246674 M * shedi what is VaVaVoom's value in sched? :)
1174246695 M * Bertl hardwire: for the use part, create a persitent tun device, assign an ip and use it inside the guest
1174246707 M * Bertl shedi: it's the calculated priority bonus a guest gets
1174246722 M * shedi I see :)
1174246730 M * Bertl hardwire: for creating, add the CAP_NET_ADMIN to the guest (and probably give 0.0.0.0 as single ip)
1174246752 M * Bertl hardwire: CAP_NET_RAW will be helpfull in most cases too
1174246921 M * hardwire Bertl: odd
1174246932 M * hardwire I wonder how ssh handles persistent tun devices
1174246991 M * Bertl check it out, and let us know
1174246996 M * hardwire MAYBE I WILL
1174247003 M * daniel_hozac Bertl: right.
1174247068 Q * hardwire Quit: Coyote finally caught me
1174247075 M * daniel_hozac Bertl: creating tun devices shouldn't require caps, IIRC
1174247110 J * hardwire ~bip@12.6.43.10
1174247112 M * daniel_hozac (but setting IP etc would)
1174247163 M * Bertl daniel_hozac: okay, so we want to check for init_pid tasks, and revaldiate them in pid1
1174247169 M * Bertl s/pid1/xid1/
1174247187 M * daniel_hozac exactly.
1174247187 M * Bertl daniel_hozac: personally, I'd prefer to revalidate all tasks in xid=1
1174247206 M * daniel_hozac makes sense.
1174247212 M * Bertl but last time that gave us quite a number of dentries
1174247301 M * Bertl OTOH, we could do some 'special' check in the lookup code
1174247309 M * Bertl for the actual pid=1 entry
1174247345 M * Bertl btw, what happens if you do a ps on the host, right before the vps?
1174247369 Q * hardwire 
1174247453 J * hardwire ~bip@12.6.43.10
1174247459 M * daniel_hozac works fine then.
1174247490 M * Bertl okay, and we also see the correct value in the host ps for pid=1?
1174247494 M * hardwire sorry
1174247504 M * daniel_hozac yeah.
1174247526 M * Bertl okay, so what we actually want is to force the xid=0 behaviour for xid=1 too, regarding init
1174247559 M * Bertl (which should solve our cosmetic issue :)
1174247585 M * daniel_hozac right
1174247638 M * hardwire Bertl: dynamic worked
1174247653 M * hardwire it just doesn't show up under ifconfig -a in a vserver context until you configure it
1174247669 M * hardwire pinging tun0 in a vserver created by ssh -w 0:0 root@vservername
1174247675 M * hardwire woot
1174247683 M * Bertl another happy user :)
1174247684 M * hardwire its a shame I didn't know this 
1174247690 M * hardwire see how I turned that around!
1174247702 M * Bertl :)
1174247726 M * hardwire however.. one little problem
1174247729 M * hardwire I can ping it
1174247731 M * hardwire it can't ping me
1174247732 M * hardwire no route :)
1174247735 M * hardwire odd eh
1174247744 M * hardwire oh hell.
1174247746 M * hardwire its broken, nm
1174247759 M * hardwire not working, I failed!! false results!!
1174247886 M * hardwire wow.. totally odd.. 
1174247892 M * hardwire bbl
1174247933 M * Bertl daniel_hozac: hmm, unfortunately it seems that is nontrivial to solve
1174247951 M * daniel_hozac Bertl: yeah, that's what i've come to think as well.
1174248107 M * Bertl let's try to get some empiric patterns ... could you add some debug info to the pid_revalidate() with as much data as possible?
1174248162 M * Bertl I think we might have a chance to work around that, by marking either the dentry or the inode as fake
1174248189 M * Bertl e.g. we could have the inode marked as init for xid=something
1174248281 M * daniel_hozac e.g. PDE->vx_flags?
1174248305 M * Bertl maybe, I'd suggest to dump everything we can reach there
1174248319 M * Bertl including the vx_proc_task_visible() state
1174248354 M * Bertl maybe we can identify some simple pattern for this
1174248381 M * daniel_hozac what inode number does a plain guest init have? same as a real init on the host?
1174248429 M * Bertl the inode numbers will not be relevant for 2.6.20+ as they are dynamic
1174248448 M * daniel_hozac oh, ok.
1174248452 M * Bertl they are only relevant for knowing that we see the _same_ inode in the revalidate
1174248484 M * Bertl i.e. we want something like %p  ... inode there too, but I guess the ino itself is not very relevant
1174248509 M * Bertl (unfortunately, that was easier with strict inode numbers)
1174249177 M * Bertl daniel_hozac: http://paste.linux-vserver.org/1326
1174249199 M * Bertl didn't we track that one down already?
1174249237 M * daniel_hozac humm, i thought so.
1174249245 M * daniel_hozac must be another issue.
1174249456 M * daniel_hozac any addr2line results?
1174249470 M * Bertl sec, machine is rebooting
1174249497 M * daniel_hozac ok.
1174249968 M * Bertl checking disks now :)
1174250428 J * matti matti@acrux.romke.net
1174250448 N * matti Guest185
1174251795 J * the-me ~me@p548A8CCD.dip0.t-ipconnect.de
1174251823 M * the-me hi, just a simple question, are the numbers of ips for a guest limited to 16 with the newest vserver patch?
1174252117 M * derjohn the-me, yep. there has always been a 16 ip per guest #define in the patch.
1174252155 M * the-me derjohn, so if I set it for example to 128 there won't be any problems?
1174252163 M * Bertl daniel_hozac: seems I have a hand today for shaking out issues for 2.2.x
1174252184 M * derjohn you can change that at compile time. The current tools (daniel's edition) should run without patch, older tools need the define at compile time, too
1174252206 J * trippeh atomt@uff.ugh.no
1174252213 M * Bertl daniel_hozac: http://paste.linux-vserver.org/1327
1174252260 M * the-me why is it still limited, if it will run fine with > 16 ips 
1174252283 M * derjohn the-me, because we like ppl to ask about that theme here in the channel </irony>
1174252288 M * Bertl the-me: because the performance for >16 ips is not optimized yet
1174252296 M * derjohn the-me, performance
1174252299 M * derjohn Bertl, race ;)
1174252311 M * the-me hehe okay, thanks ;)
1174252340 M * derjohn the-me, i ran a guest with >16 and <64 IP ... 
1174252358 M * trippeh So, util-vserver 0.30.212 is too old for 2.6.20.3-vs2.2.0-rc18? Getting "ncontext: vc_net_create(): Invalid argument" when starting a context
1174252385 M * Bertl trippeh: nope, it works quite fine, just a config issue
1174252398 M * trippeh Hmm
1174252429 M * Bertl trippeh: i.e. you have to enable certain legacy options and such (in the kernel) to make it work with older tools
1174252458 M * Bertl trippeh: but actually you want to switch to new tools and new config with fixed context ids and such (especially the latter one)
1174252539 M * trippeh CONFIG_VSERVER_LEGACY? Hmm. I think I'd rather just upgrade the tools :)
1174252591 M * Bertl check the logs (especially dmesg) it probably contains related info
1174252602 M * trippeh dynamic contexts disabled.
1174252603 M * trippeh Hmm.
1174252649 M * Bertl which is deprecated for more than two years now, IIRC .... echo 42 >/etc/vservers/<name>/context 
1174252663 M * Bertl context ids are between 2 and 49151 for now
1174252672 M * Bertl and they should be unique for each guest
1174252701 M * trippeh Aha. I better kick this debianish wrapper thingy then, for not allocating a context id.
1174252712 M * trippeh On create
1174252728 M * Bertl yes, that is a good idea ... and file a bug report too
1174252746 Q * the-me Remote host closed the connection
1174252766 M * trippeh It starts now :)
1174252774 M * daniel_hozac should fix itself automatically once you use util-vserver 0.30.213...
1174252787 M * trippeh daniel_hozac: Is it released yet?
1174252794 M * trippeh www.linux-vserver.org says 212
1174252808 M * daniel_hozac should be released in an hour or so...
1174252814 M * trippeh Ha :-)
1174252823 M * Bertl rc5 is already out there
1174252895 M * derjohn Bertl, trippeh file a bug? in util-vserver for not auto-allocating a context id in userspace ?
1174252917 M * Bertl daniel_hozac: the second oops seems to be related to the unhashing we probably fixed in 2.6.20 only
1174252927 M * derjohn daniel_hozac, 213 will have userspace generated context-IDs ???
1174252942 M * Bertl derjohn: no, for the debian wrapper existing and still using dynamic contexts 
1174252942 M * daniel_hozac derjohn: yes.
1174252966 N * DoberMann DoberMann[ZZZzzz]
1174253025 M * trippeh the Debian-wrapper "newvserver" just calls vserver build, so I guess it will fix itself with 213
1174253062 M * Bertl it is still quite useless
1174253074 M * derjohn trippeh, ah. understood. I never used that thingy. i didnt even know that it exists.
1174253115 M * trippeh Bertl: That may be :)
1174253224 Q * DreamerC Ping timeout: 480 seconds
1174253312 M * DavidS trippeh: i never used newvserver from debian, but then Bertl had a long talk with me about the debian packages ;))
1174253343 M * daniel_hozac and now you're its biggest fan? :)
1174253394 M * Bertl yeah, you have to have a reason for that :)
1174253532 M * DavidS trippeh: i never used newvserver from debian, but that's probably because Bertl had a long talk with me about the debian packages ;))
1174253534 M * DavidS better
1174253542 M * trippeh Only ten more vservers to manually assigns id's to, before booting 2.2.0-rc18 in production ;)
1174253544 A * DavidS should learn english
1174253599 M * trippeh I'm not to fond of the debian vserver packages either, but they seem to be working, and I'm lazy ;)
1174253861 Q * derjohn Quit: by(t)e
1174253915 M * trippeh Lazy wins
1174254041 M * DavidS indeed .. that's why i'm still waiting for 2.2 to appear in the debian stock kernel ... ;)
1174254152 Q * trippeh Quit: remote console is for wussies, brb
1174254176 M * Bertl DavidS: will be there any year now :)
1174254276 M * daniel_hozac probably :)
1174254305 M * DavidS I'm very sure that debian will release before may.
1174254319 M * DavidS (please don't ask me which year ;)
1174254354 M * daniel_hozac hehe.
1174254358 M * DavidS When will there be a stable 2.2 (if we are already on this topic)?
1174254371 M * daniel_hozac once the bugs are fixed :)
1174254378 M * Wonka i got told debian will release at April 2nd, and that date is not to be changed again
1174254380 M * daniel_hozac 3 open issues now?
1174254403 M * DavidS daniel_hozac: that sounds like debian's WIR, SIYH ;)
1174254434 M * daniel_hozac what?
1174254438 M * Bertl daniel_hozac: we are missing the delta-unhash-fix01.diff for 2.6.19 and IMHO it is incomplete ...
1174254473 M * daniel_hozac isn't that missing from 2.6.20 as well?
1174254486 J * trippeh atomt@x.vx.no
1174254499 M * daniel_hozac but yes, that got lost in the 2.6.19 port IIRC.
1174254528 M * daniel_hozac how is it incomplete though?
1174254667 M * trippeh Now, if I only had connected the serial console when I was on-site yesterday ;)
1174254702 A * trippeh gives it another 5 minutes
1174254777 M * Bertl daniel_hozac: I think we have to release() the vxinfo in the error path
1174254792 M * Bertl i.e. as the __create() now claims the vxinfo
1174254792 M * daniel_hozac which one?
1174254866 M * Bertl when vc_ctx_create() fails, we have to release that too, IMHO
1174254884 M * daniel_hozac right, don't we?
1174254903 M * daniel_hozac isn't that what the release_vx_info(new_vxi, NULL) right after out: does?
1174254938 M * Bertl yes, that would be fine
1174254962 M * Bertl ah, yes, it's part of the patch
1174254977 M * Bertl the real issue is, that neither 2.6.19 nor 2.6.20 has it! :)
1174254987 M * daniel_hozac indeed.
1174255017 A * Bertl was looking at the 2.6.20 code, assuming it was included there
1174255041 M * daniel_hozac no, as i said, i believe it was lost in the 2.6.19 port.
1174255055 M * Bertl okay, I will add it back and retest
1174255084 M * Bertl minor issue with the WARN option
1174255089 M * Bertl In file included from kernel/vserver/context.c:46:
1174255089 M * Bertl kernel/vserver/cvirt_init.h: In function 'vx_info_exit_cvirt':
1174255089 M * Bertl kernel/vserver/cvirt_init.h:47: warning: unused variable 'value'
1174255138 M * daniel_hozac i guess that's always been there, if you've disabled debugging.
1174255181 M * DavidS When will Debian release? When It's Ready; Sooner If You Help.           a standard quip on #debian
1174255249 M * daniel_hozac i'm a big fan of the deadline release schemes, except they don't work very well when it's not your day job...
1174255303 M * daniel_hozac i still haven't had time to thoroughly test all of the 0.30.213 features, and i'd hate to do a brown paper bag release.
1174255340 M * Wonka i'd help, if i had a machine...
1174255439 M * daniel_hozac this is of course an area where automation could help. but that also requires time ;)
1174255442 M * Bertl daniel_hozac: they don't work very well. period.
1174255479 M * daniel_hozac hehe
1174255484 M * DavidS the only deadline that works is "tomorrow" .. and then only for very small work packages
1174255517 M * Bertl having a deadline is useful, if you can push it without any problems ...
1174255543 M * Bertl the trick is that you need to convince yourself that it _is_ a deadline, up to the moment when you push it :)
1174255565 J * renihs ~penguin@83-65-34-34.arsenal.xdsl-line.inode.at
1174255572 M * Bertl wb renihs!
1174255573 M * DavidS tell me .. I should be writing my masters thesis ... 
1174255593 M * DavidS instead i'm fiddling with my vservers :-P
1174255605 M * Bertl DavidS: do they need attention?
1174255620 M * daniel_hozac ugh, externalize is still broken.
1174255793 M * DavidS no, I'm just improving my puppeteer skills and am slowly converting everything into automated recipes ... today i doubled the number of nagios tests by automating DNS zone distribution and adding nagios checks for each
1174255922 M * Bertl DavidS: okay, was worried that something is not working as expected
1174255940 J * cdrx ~legoater@cap31-3-82-227-199-249.fbx.proxad.net
1174255943 M * Bertl DavidS: you ahve to apologize, but usually folks show up here when something is not working as expected :)
1174255948 M * Bertl evening cdrx!
1174255962 M * Bertl (hmm, probably afternoon :)
1174255991 M * cdrx well, it's close to bed time for me 
1174256132 M * daniel_hozac Bertl: okay, externalize should work now (read: works here) in trunk.
1174256148 M * Bertl okay, rc6?
1174256170 M * DavidS Bertl: no prob ... in a sense something is not working, but that's probably nothing you can fix  *sigh*
1174256208 M * daniel_hozac just script changes, so it shouldn't be necessary (i.e. patch should work fine).
1174256223 M * Bertl okay, url for the patch?
1174256299 M * Bertl btw, how would I go about adding a script (to be run after install) for a template based guest/install?
1174256327 M * Bertl is there some way to 'hide' the template behind a distro specification?
1174256363 M * daniel_hozac i.e. to not specify the tarball on the command line?
1174256365 M * Bertl i.e. that I could do -m template -d mdv2007 and it knows what scripts and what tar to use?
1174256396 M * daniel_hozac not presently, but i suppose you might use an initpost script to do the untarring yourself.
1174256434 M * daniel_hozac http://people.linux-vserver.org/~dhozac/p/uv/experimental/delta-rc5-trunk.diff
1174256446 M * Bertl is something like that planned?
1174256456 M * daniel_hozac it certainly makes sense.
1174257397 M * trippeh Argh, it was a kernel OOPS. Too long for the console, and the serial cable was not connected, plus it doesn't happen again
1174257409 M * daniel_hozac where was it?
1174257439 M * trippeh Only got the last few lines of the oops. So not sure.
1174257452 M * trippeh It happened during boot though. Came up clean after a reset.
1174257466 M * daniel_hozac last line should have EIP these days.
1174257475 M * daniel_hozac (for precisely that reason)
1174257491 M * trippeh Then it didn't get all the way  either ;)
1174257507 M * trippeh Even sysrq was dead
1174257564 M * trippeh It did get to mounting the file system, but not far enough to actually get the OOPS to disk
1174257570 M * trippeh systems, even
1174257589 A * trippeh goes looking for a serial cable ;)
1174257715 M * daniel_hozac well, i'm really tired and i have to get up early tomorrow. as i'm not done testing 0.30.213, it'll have to wait a bit longer... good night folks!
1174257841 M * Bertl reg. 1326: include/linux/tty.h:341
1174257880 M * Bertl triggered from fs/proc/array.c:421
1174257923 M * Bertl I'm opening a bug hunt entry for that
1174257935 M * daniel_hozac ah, it's the tty_devnum bug in disguise?
1174257953 M * Bertl did we resolve it? IIRC, no ...
1174257957 M * daniel_hozac no.
1174258133 J * Aiken ~james@ppp250-73.lns2.bne4.internode.on.net
1174258137 M * Bertl morning Aiken!
1174258142 M * Aiken hi
1174258151 M * Bertl daniel_hozac: for me that looks like tty->driver is NULL
1174258188 M * phreak`` daniel_hozac: seems to work so far :)
1174258299 M * daniel_hozac Bertl: yeah, i agree.
1174258349 M * daniel_hozac but, i'm really off to bed now, good luck!
1174258356 Q * trippeh Quit: tjohei
1174258370 J * trippeh_ atomt@uff.ugh.no
1174258375 N * trippeh_ trippeh
1174258394 M * Bertl daniel_hozac: have a good one!
1174258433 M * trippeh Ok, back to previous kernel for now. Failed to locate a serial cable, so I'll better just try to reproduce in lab instead ;)
1174258550 M * Bertl Hollow, phreak``: ping?
1174258560 M * Hollow pong?
1174258589 M * Bertl we 'just' discovered that we accidentially dropped a fix regarding hashing/unhashing form 2.2.0
1174258611 M * Bertl IIRC, this was the solution to some issues the persistant contexts triggered
1174258623 M * Bertl here is a patch which adds that back to 2.2.0
1174258627 M * Bertl http://vserver.13thfloor.at/Experimental/delta-unhash-fix02.diff
1174258641 M * Bertl could you test that with one of your setups ASAP?
1174258671 M * Hollow yep, will look into it tomorrow
1174258678 M * Bertl excellent! tx!
1174258687 M * phreak`` Bertl: got a testcase for that ?
1174258697 M * Hollow have to setup my test environment again on my new box
1174258720 M * Bertl phreak``: I did hit it when spawning quite a number of contexts
1174258739 M * phreak`` quite a number >= 2 ?
1174258740 M * phreak`` :D
1174258755 M * Bertl phreak``: but the interesting part is probably the persistant case with and without a failing helper
1174258764 M * Bertl phreak``: 800+
1174258782 M * phreak`` uh
1174258787 M * phreak`` 800 contexts ? :)
1174258800 M * phreak`` Bertl: I think matti was right .. you are an alien :)
1174258800 M * Bertl yes :)
1174258824 M * phreak`` Bertl: just empty contexts or full-blown guests ?
1174258850 M * Bertl well, if you fully populate the context, it was hit around 80 or so (without any sharing on a moderate machine)
1174258883 M * Bertl but I think it is easier to trigger with negative helper results
1174258888 M * phreak`` (I still haven't converted my box to anything > 2.6.18 :P)
1174258916 M * Bertl would be a good time to do that now, for testing
1174258925 M * Bertl before we release 2.2.0 :)
1174258959 M * Bertl okay, off for now ... back a little later ...
1174258962 M * phreak`` hrm, well the libata stuff makes it quite hard :)
1174258968 M * phreak`` and I'm going to bed :P
1174258971 N * Bertl Bertl_oO
1174259355 Q * arachnist Ping timeout: 480 seconds
1174260010 J * DreamerC ~dreamerc@125-225-105-11.dynamic.hinet.net
1174260063 Q * FireEgl Quit: ...
1174260368 M * trippeh I just did two, one failed.
1174260386 M * trippeh Not even sure if it was in vserver code, silly serial cable thieves!
1174260493 J * arachnist ~arachnist@088156185052.who.vectranet.pl
1174260938 Q * ema Quit: leaving
1174261532 Q * meandtheshell Quit: Leaving.
1174262215 Q * boci^ Quit: Távozom