1180224726 M * Bertl okay, off to bed now .. have a good one everyone! cya!
1180224733 N * Bertl Bertl_zZ
1180225306 J * besonen_mobile ~besonen_m@71-220-233-253.eugn.qwest.net
1180226772 Q * Piet Quit: Piet
1180227656 Q * Hurga Remote host closed the connection
1180230863 J * DoberMann_ ~james@AToulouse-156-1-122-59.w90-30.abo.wanadoo.fr
1180230965 Q * DoberMann[ZZZzzz] Ping timeout: 480 seconds
1180239150 J * zLinux_ ~zLinux@88.213.15.8
1180239513 Q * zLinux[] Ping timeout: 480 seconds
1180240449 Q * duckx Ping timeout: 480 seconds
1180242496 P * zLinux_ Leaving
1180242602 J * duckx ~Duck@tox.dyndns.org
1180243172 Q * duckx Ping timeout: 480 seconds
1180247853 Q * Johnnie Read error: Connection reset by peer
1180247957 J * Johnnie ~jdlewis@c-67-163-247-109.hsd1.pa.comcast.net
1180248675 M * mattzerah Bertl_zZ, ping
1180249229 J * dna ~naucki@177-198-dsl.kielnet.net
1180251219 J * duckx ~Duck@tox.dyndns.org
1180251607 Q * mire Quit: Leaving
1180251643 J * mire ~mire@14-170-222-85.adsl.verat.net
1180251989 Q * duckx Ping timeout: 480 seconds
1180252612 J * mattzerah_ ~matt@121.50.222.55
1180252634 N * mattzerah mattzerah__
1180252646 N * mattzerah_ mattzerah
1180252850 Q * mattzerah 
1180252857 Q * mattzerah__ Quit: mattzerah__
1180252876 J * mattzerah ~matt@121.50.222.55
1180253802 Q * dna Quit: Verlassend
1180255610 J * duckx ~Duck@tox.dyndns.org
1180256158 Q * duckx Ping timeout: 480 seconds
1180257193 J * _are_ ~are@dslb-084-057-185-096.pools.arcor-ip.net
1180257196 M * _are_ Hi
1180257314 J * DavidS ~david@p57a4c78b.dip0.t-ipconnect.de
1180257371 M * DavidS yay: Debian/sid/linux-2.6/2.6.21-1:   * Update vserver patch to 2.2.0-rc1.
1180257767 J * bonbons ~bonbons@ppp-110-113.adsl.restena.lu
1180257804 J * duckx ~Duck@tox.dyndns.org
1180258324 Q * duckx Ping timeout: 480 seconds
1180258489 Q * _are_ Remote host closed the connection
1180259433 J * dna ~naucki@220-196-dsl.kielnet.net
1180259776 J * mattzerah_ ~matt@121.50.222.55
1180259781 P * mattzerah_ 
1180260063 J * duckx ~Duck@tox.dyndns.org
1180261857 N * ag- Guest450
1180261879 N * Guest450 ag-
1180263196 N * DoberMann_ DoberMann
1180265019 N * Bertl_zZ Bertl
1180265032 M * Bertl morning folks!
1180265051 M * daniel_hozac morning Bertl!
1180265056 M * Bertl mattzerah: pong!
1180265471 Q * Aiken Quit: Leaving
1180266366 Q * dna Quit: Verlassend
1180267491 J * Piet hiddenserv@tor.noreply.org
1180268474 Q * DavidS Quit: Leaving.
1180271132 M * Bertl okay, off for now ... should be back in a few hours
1180271178 N * Bertl Bertl_oO
1180274301 J * arachnis1 arachnist@088156189068.who.vectranet.pl
1180274301 Q * arachnist Read error: Connection reset by peer
1180274314 N * arachnis1 arachnist
1180278758 Q * phedny Read error: Connection reset by peer
1180278782 J * phedny ~mark@ip56538143.direct-adsl.nl
1180279258 Q * eyck Quit: leaving
1180279670 J * eyck ~eyck@nat.nowanet.pl
1180281202 J * |jmcaricand| ~kvirc@d83-179-195-51.cust.tele2.fr
1180281378 J * dna ~naucki@146-230-dsl.kielnet.net
1180281836 N * Bertl_oO Bertl
1180281846 M * Bertl okay, I'm back ..
1180282749 J * Chainsaw ~gangster@attic.vleuten.org
1180283010 J * the-me ~me@p548AB078.dip0.t-ipconnect.de
1180283038 M * the-me hi, is the 2.2.0 vserver patch working with 2.6.21.3?
1180283079 J * Johnsie ~jdlewis@c-67-163-247-109.hsd1.pa.comcast.net
1180283085 M * Chainsaw heya
1180283087 M * Chainsaw questionm
1180283102 M * Chainsaw trying to get honeyd working in vservers,why am is the operation not permitted to add routes
1180283111 M * Chainsaw do i need to change something in my config of vserver?
1180283255 Q * Johnnie Quit: G'bye!
1180283266 Q * Johnsie 
1180283275 J * Johnnie ~jdlewis@c-67-163-247-109.hsd1.pa.comcast.net
1180284097 Q * |jmcaricand| Quit: KVIrc 3.2.4 Anomalies http://www.kvirc.net/
1180284099 N * DoberMann DoberMann[PullA]
1180286157 M * mjt Chainsaw: changing routes from within a guest isn't permitted by default
1180286163 M * mjt and it's the Right Thing (tm) :)
1180286181 M * mjt the best is to configure needed routing on the host
1180286665 M * Chainsaw yarr
1180286676 M * Chainsaw you are familiar with honeypots?
1180286727 M * mjt not really, but i at least imagine what's the talk about ;)
1180286736 M * Chainsaw well
1180286747 M * Chainsaw honeynets need a different network to operate in
1180286758 M * Chainsaw because it will operate on the net as a 'open' box
1180286765 M * mjt sure
1180286766 M * Chainsaw it will simulate services,ports open etc
1180286788 M * Chainsaw so i think it's best to run in the virtual itself,als the routes
1180286801 M * mjt not necessary "networks", but it's best to have at least separate IP address(es) for them
1180286809 M * Chainsaw yar
1180286828 M * Chainsaw im gonna have another cup and check it out,thnx for the input ;)
1180286834 M * Chainsaw what you up to b.t.w?
1180286840 M * mjt you understand that in vserver, *all* networking comes from the host, right?
1180286867 M * mjt here we've only one kernel, and it's the kernel that does networking
1180286959 M * Chainsaw yeah
1180286962 M * Chainsaw i do
1180287026 M * mjt now if you tell me how honeyd works on a regular linux box (how it receives network packets and how networking setup looks like), i'll probably can tell you how to do the same in vserver.
1180287052 M * Chainsaw but that would take the fun away!
1180287053 M * Chainsaw :)
1180287064 M * mjt alternatively, you can grant your vserver an ability to change routes and what not - but obviously it's not recommended.
1180287120 M * Chainsaw ans
1180287122 M * Chainsaw and
1180287125 M * Chainsaw the vserver
1180287128 M * Chainsaw runs in vmware ;)
1180287157 M * mjt well, nothing's wrong with that.  Not very useful for production, but good for experiments.
1180287890 Q * Piet Quit: Piet
1180288693 J * onox ~onox@kalfjeslab.demon.nl
1180290557 Q * Chainsaw 
1180293080 Q * onox Quit: leaving
1180294406 J * meandtheshell ~markus@85-124-233-82.work.xdsl-line.inode.at
1180294962 Q * meandtheshell Quit: Leaving.
1180295203 J * Piet hiddenserv@tor.noreply.org
1180295376 J * meandtheshell ~markus@85-124-37-63.dynamic.xdsl-line.inode.at
1180297206 J * tuxmania ~bonbons@158.64.111.187
1180297517 Q * bonbons Ping timeout: 480 seconds
1180297628 J * ema ~ema@rtfm.galliera.it
1180297754 Q * nou Ping timeout: 480 seconds
1180297765 N * tuxmania bonbons
1180297922 J * nou Chaton@causse.larzac.fr.eu.org
1180298505 Q * trippeh Ping timeout: 480 seconds
1180298596 J * trippeh atomt@uff.ugh.no
1180299834 J * Aiken ~james@ppp233-152.lns1.bne4.internode.on.net
1180300119 J * Hollow ~hollow@styx.xnull.de
1180300439 J * tuxmania ~bonbons@ppp-111-204.adsl.restena.lu
1180300759 Q * bonbons Ping timeout: 480 seconds
1180301812 N * tuxmania bonbons
1180303173 M * Bertl daniel_hozac: the kernel issues are still in latest 2.6.21.3, right? or was there a fix for it yet?
1180303179 Q * Piet Ping timeout: 480 seconds
1180303184 M * Bertl s/kernel/sysfs/
1180303219 M * bonbons Bertl: as far as I saw on LKML there is no fix yet
1180303241 M * Bertl great ...
1180303312 M * Bertl maybe it becomes a 'feature' soon :)
1180303328 M * daniel_hozac hehe
1180303343 M * Bertl do we know what triggers it by now?
1180303356 M * Bertl or is it still something which happens every now and then?
1180303778 M * daniel_hozac would ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/2.6.22-rc1/2.6.22-rc1-mm1/broken-out/gregkh-driver-sysfs-allocate-inode-number-using-ida.patch be the same problem, or a different one?
1180303979 M * Bertl looks good to me
1180303997 M * Bertl we probably should include that into the vserver patches for now
1180304166 Q * dna Quit: Verlassend
1180304507 M * Bertl but I guess that patch requires quite a number of other patches
1180304755 M * daniel_hozac http://lkml.org/lkml/2007/5/21/598 appears to be a backport.
1180305163 M * Bertl btw, I just realized that sysfs is deprecated now?
1180305199 M * Bertl wasn't sysfs _created_ because filesystem interfaces are so much better than syscall interfaces?
1180305230 M * sid3windr sysfs deprecated? o_O 
1180305251 M * daniel_hozac hmm?
1180305254 M * daniel_hozac what makes you say that?
1180305338 M * Bertl Create deprecated sysfs files (SYSFS_DEPRECATED) [Y/n/?] (NEW) n
1180305357 M * Bertl (from 2.6.21.3 :)
1180305388 M * sid3windr that's a general option? or sysfs for just one driver/part ?
1180305402 Q * the-me Remote host closed the connection
1180305405 M * mnemoc sysfs deprecated??
1180305458 M * daniel_hozac Bertl: that's just for some files.
1180305480 M * Bertl ah, okay then, sorry for the noise ...
1180305705 M * sid3windr :)
1180305712 M * sid3windr TSKTSK! ;
1180305713 M * sid3windr )
1180305746 M * Bertl btw, I just realized that 2.6.19.7 claims to do NCQ for most sata controllers, but actually doesn't do it :)
1180305782 M * Bertl so if somebody with a fast SATA disk/host/controller system is wondering why the latencies are sooo high ...
1180305896 M * daniel_hozac is that fixed in more recent kernels?
1180305911 M * Bertl I don't know yet ... but I will know soon :)
1180305933 M * arachnist May  2 11:59:47 nest ata1.00: ATA-7, max UDMA/133, 625142448 sectors: LBA48 NCQ (depth 0/32)
1180305942 M * Bertl that means not working :)
1180305955 M * Bertl depth > 0 means working :)
1180305985 M * Bertl max for depth is 31, because of an ATA specification bug
1180306023 M * Bertl arachnist: if you want to get funny messages from your kernel, you can try:
1180306102 M * Bertl echo 31 > /sys/block/sdX/device/queue_depth
1180306127 M * Bertl but on 2.6.19.x, that usually gives a kernel oops :)
1180306137 M * arachnist zsh: permission denied: /sys/block/sda/device/queue_depth
1180306149 M * Bertl ah, intel AHCI?
1180306151 M * arachnist nah
1180306161 M * arachnist nforce550
1180306245 J * yarihm ~yarihm@84-75-103-239.dclient.hispeed.ch
1180306339 M * Bertl I also saw that there is a fault injection framework in 2.6.21 ... can probably be removed again in a few kernel vertsions from now :)
1180306347 M * Bertl *versions
1180306388 M * daniel_hozac hehe.
1180306503 Q * bonbons Quit: Leaving
1180307007 J * olli ~opera@85-220-112-127.dsl.dynamic.simnet.is
1180307052 M * olli Hi
1180307065 M * olli anyone active in this room?
1180307073 M * mnemoc *cough*
1180307078 M * olli *g*
1180307080 M * olli cool
1180307087 M * Bertl nope, all passively waiting for your question :)
1180307093 A * mugwump_ gives mnemoc a tissue
1180307099 N * mugwump_ mugwump
1180307116 M * olli Alright, I got a question about how to limit the resources, especially hard disk space for a vserver
1180307120 M * Bertl daniel_hozac: shouldn't the coretemp driver be in 2.6.21 too?
1180307166 M * olli My first attempt was to mount a monolithic file (via /dev/loopX) on the directory supposed to be the vservers /
1180307187 M * Bertl brutal, but effective
1180307198 M * olli hmm, didn't work for me
1180307214 M * Bertl how so?
1180307229 M * olli no idea
1180307238 M * Bertl I mean, a filesystem is a filesystem, it will not resize without good reason :)
1180307253 M * olli once I stopped the server I saw that the directory was indeed in use, but the mounted monolithic file wasn't touched :-|
1180307280 M * Bertl sounds like you did something wron then
1180307296 M * olli well that's where my question comes in ;)
1180307309 M * s0undt3ch hello ppl
1180307310 M * olli could it be that a bind-mount lets this fail
1180307312 M * olli hi
1180307316 M * s0undt3ch clarify something to me
1180307345 M * Bertl olli: why would you --bind mount if you have a loop device anyway?
1180307365 M * s0undt3ch does the mac address of a vserver's guest eth stays static or it changes at every guest restart?
1180307366 M * olli because my /var is actually a bind-mount to a directory on my data partition 
1180307434 M * Bertl s0undt3ch: mac is layer2, Linux-VServer is layer 3 (for now :)
1180307458 M * s0undt3ch Bertl: ok, sorry for my ignorance, and? :)
1180307480 M * Bertl s0undt3ch: the mac is host specific, the guest will never touch it
1180307510 M * s0undt3ch so the mac address will be equal for all guests?
1180307524 M * s0undt3ch equal to the host's mac address?
1180307533 M * daniel_hozac yes.
1180307562 M * s0undt3ch anyway, nevermind, this is a stupid question, guests won't have to get address from dhcp :)
1180307594 M * s0undt3ch I'm in the middle of configuring my wireless router, hence these questions
1180307597 M * s0undt3ch Thanks anyway
1180307694 M * Bertl s0undt3ch: actually the guest _could_ get the address via dhcp
1180307707 M * Bertl but they would need to provide a special identifier
1180307763 M * s0undt3ch well, I'm moving from static to router's dhcp setup, and the guests don't need to quesry dhcp, I just need to make sure their addresses are not assigned to others that's all
1180307840 M * olli Bertl: presuming I'd fix the issue (since you said it is supposed to work with a file mounted through loop), can I simply move/copy the contents of the directory into the mounted file lateron?
1180307912 M * Bertl yes, but as I already implied, the loopback is a very brutal method to accomplish what you are looking for
1180307938 M * Bertl a better approach would be to use an lvm partition for each guest
1180307945 M * Bertl (less overhead)
1180307953 M * daniel_hozac and the best would be a shared filesystem with disk limits
1180307964 Q * ema Quit: leaving
1180307967 M * Bertl and the best solution would be to use a tagged filesystem for all guests with limits as daniel_hozac stated
1180308009 M * olli alright, how do I implement the latter one
1180308024 M * olli the documentation on the project site appears to be a little arcane :-D
1180308085 M * Bertl it's quite simple, mount the filesystem in question with the tagxid option
1180308105 M * olli okay
1180308108 M * Bertl configure the maximum allowed space in /etc/vservers 
1180308113 M * daniel_hozac http://oldwiki.linux-vserver.org/Disk+limits
1180308126 M * Bertl make sure that the initial files are tagged properly, and that's it
1180308140 M * olli ah, nice
1180308149 M * olli that sounds cool
1180308157 M * olli thanks
1180308213 M * olli the parameter after -c is the name/context of the vserver, I presume?
1180308232 M * Bertl most likely (no idea what command :)
1180308232 M * olli does this affect the vunify feature - e.g. hardlinking not working anymore or so?
1180308255 M * Bertl no, that works quite fine, just unified files will not be credited to any host
1180308265 M * Bertl s/host/guest
1180308277 M * olli I see
1180308292 M * olli so only once the link breaks?
1180308297 M * Bertl yep
1180308307 M * olli alright, thanks a lot :)
1180308314 M * Bertl you're welcome! 
1180308505 M * daniel_hozac note that if you have already hashified/unified your guests, you'll want to use the chxid -U.
1180308512 M * daniel_hozac (which skips those files)
1180308928 M * olli thanks for the hint
1180308934 M * olli haven't done that yet, though
1180308947 M * olli however, another problem turned up now
1180308976 M * olli "vcontext: chroot(): Permission denied"
1180308998 M * olli and ...
1180309000 M * olli An error occured while executing the vserver startup sequence; when
1180309000 M * olli there are no other messages, it is very likely that the init-script
1180309000 M * olli (/etc/init.d/rc 3) failed.
1180309104 M * olli does this refer to the host or the guest startup sequence?
1180309347 M * daniel_hozac guest, of course.
1180309362 M * daniel_hozac what kernel are you using?
1180309508 M * daniel_hozac or, ah, did you tag your files now? what does lsxid -d /vservers/<guest> show?
1180309514 N * DoberMann[PullA] DoberMann[ZZZzzz]
1180309614 M * olli yes, tagged them
1180309662 M * olli Well, apparently that was the problem: "!!ERR!!               /data/var/lib/vservers/asgard"
1180309776 M * daniel_hozac you didn't mount the filesystem with tagxid?
1180309818 M * olli yes
1180309830 M * olli with chxid I have to give the name or the numeric ID?
1180309836 M * daniel_hozac so grep tagxid /proc/mounts shows it, yes?
1180309839 M * daniel_hozac either works.
1180309884 M * olli aaah
1180309885 M * olli okay
1180309889 M * olli my bad
1180309913 M * olli yes, shows tagxid
1180309937 M * olli the problem was the static ID
1180309946 M * daniel_hozac oh?
1180309951 M * olli hah
1180309954 M * olli works now ;)
1180309960 M * daniel_hozac i.e. you didn't have one?
1180309967 M * olli *g* yepp
1180309969 M * olli sorry ;)
1180309986 M * daniel_hozac okay then.
1180310006 M * olli Thanks a thousand for helping me track it down
1180310025 M * olli and also thanks for the link to the old wiki ;)
1180310030 M * daniel_hozac you're welcome!
1180310155 M * olli one last question
1180310174 M * olli is there any way to "debug" problems with "vserver <name> enter"?
1180310213 M * daniel_hozac what kind of problem would that be?
1180310227 M * olli "bash: 7: Bad file descriptor"
1180310246 M * olli I presume it has to do with chxid, since I haven't done anything else there
1180310256 M * daniel_hozac that's... interesting.
1180310313 M * daniel_hozac could you strace it?
1180310320 M * daniel_hozac see what file descriptor 7 is, if anything?
1180310340 M * olli let me see