1158711604 Q * cehteh Quit: Reconnecting to server - dircproxy 1.0.5
1158711615 J * cehteh ~ct@pipapo.org
1158712047 J * AjAx-- hiddenserv@tor.noreply.org
1158712095 Q * AjAx- Ping timeout: 480 seconds
1158712104 N * AjAx-- AjAx-
1158712660 N * Bertl_oO Bertl
1158712672 M * Bertl evening folks!
1158712732 M * Bertl |yang|: 2.6.17 vanilla?
1158713140 M * matti Hello Bertl :)
1158713149 M * Bertl hey matti!
1158713564 M * matti What's up?
1158713594 M * Bertl everything fine here so far, and for you?
1158713661 M * matti Fine so far, thanks :-)
1158714411 M * id23 n8 to Bertl and the other nonsleepers 
1158714809 M * Bertl hehe
1158714926 Q * id23 Ping timeout: 481 seconds
1158716547 Q * cehteh Ping timeout: 480 seconds
1158717180 M * matti :)
1158717274 Q * fluor Ping timeout: 480 seconds
1158717401 M * matti k, time to sleep.
1158717403 M * matti Night Bertl :)
1158717406 M * Bertl good night!
1158717417 M * matti ;]
1158717890 J * AjAx-- hiddenserv@tor.noreply.org
1158718057 M * Radiance for some reason quota's refuse to work on the vps, despite they're enabled and active on the host, and in the vps itself also in fstab
1158718074 M * Radiance any one struggled with strangling quota nastiness ?
1158718108 M * Bertl well, usually it is some kind of misconfiguration
1158718128 M * Bertl but first, what filesystem, kernel and tool versions are we talking about?
1158718156 Q * AjAx- Ping timeout: 480 seconds
1158718160 M * Radiance lemme get that for you mate 
1158718169 N * AjAx-- AjAx-
1158718290 J * mire ~mire@102-167-222-85.COOL.ADSL.VLine.verat.net
1158718346 M * Radiance BertI, 2.6.17.7, 0.30.210, vs2.0.2-rc2, host filesystem is ext3, vps fs is ufs
1158718384 M * Radiance (this problem existed with previous versions too for me, but i hadn't much time to look into it)
1158718398 M * Bertl okay, you did mount the filesystem on the host or is it guest only?
1158718418 M * Radiance yes, in the host it's mounted and has quota's enabled
1158718503 M * Radiance -rw-------   1 root root  13312 2006-09-17 06:33 aquota.group
1158718503 M * Radiance -rw-------   1 root root  12288 2006-09-20 01:00 aquota.user
1158718517 M * Radiance this is found in the /var (which is the partition where the vserver is located)
1158718581 M * Radiance within the vps itself...would it be required to also create these files in the root ?
1158718597 M * Bertl no, it doesn't work this way
1158718613 M * Bertl i.e. you cannot put only a subdirectory of a filesystem into a guest
1158718621 M * Bertl and use quotas from within
1158718623 M * Radiance ah ok
1158718638 M * Bertl i.e. you have to have the mountpoint inside the guest
1158718649 M * Bertl so that the tools inside can access those special files
1158718681 M * Radiance i see, now i'm thinking how to adapt to that heh
1158718702 J * Aiken_ ~james@tooax6-103.dialup.optusnet.com.au
1158718763 M * Bertl well, you have only one guest?
1158718767 M * Radiance yes
1158718827 M * Radiance i hope i don't have to assign that whole partition to the guest ?? hehe
1158718870 M * Bertl well, probably the best solution would be to create a separate partition for that guest
1158719007 M * Radiance hmm, so if i want to use quota's then there is no other way then to create a partition for each guest ? 
1158719024 M * Bertl yes, and no
1158719028 Q * Aiken Ping timeout: 480 seconds
1158719036 M * Radiance ah, or have the guest sit in /mountpoint
1158719039 M * Bertl basically, we have a way to allow for shared quota
1158719067 M * Radiance (which is in this case not possible in /var)
1158719072 M * Bertl but this is not finished (because very low number of requests and nobody to test it) for the 2.6 kernel
1158719091 M * Bertl you can get and use it with the 2.4 kernel though :)
1158719118 M * Bertl without that 'feature' you have to put the entire fs into the guest
1158719131 M * Bertl you can still share it with other guests or the host
1158719160 M * Bertl btw, what do you need the quota for?
1158719202 M * Radiance so if understand correctly, for quota's the guest must be able to access the files in the root of the mountpoint so it has to be a partition
1158719244 Q * gerrit_ Ping timeout: 480 seconds
1158719246 M * Radiance i would like to use it to limit users on how much they can upload, it's for a non profit club
1158719266 M * Bertl upload via ftp? ssh? sftp? web?
1158719272 M * Radiance ssh
1158719281 M * Radiance well the sftp part of ssh
1158719317 M * Bertl okay, in this case it would suffice to set the limits on the host, no?
1158719341 M * Radiance well they're in the vps, and right now it's not possible the way it's been setup :(
1158719368 M * Bertl what I mean is, you should have no need to configure it from _inside_ the guest, no?
1158719391 M * Radiance the host is offlimits to the guy i try to help
1158719404 M * Radiance that's why the vps too
1158719422 M * Bertl well, you said: 04:27 < Radiance> _i_ would like to use it to limit users on how much they can upload
1158719449 M * Bertl now it is _the guy_ inside the guest?
1158719455 M * Radiance yes, cause I <-- i'm the one who is now trying to get this going so i can focus on my other stuff :)
1158719505 M * Bertl so is it required that _he_ does the limiting?
1158719569 M * Radiance i'm not getting paid by them to administratate their users and such, just trying to help a hand, you know...heart to heart, friend to friend :)
1158719572 Q * Johnnie Remote host closed the connection
1158719604 M * Bertl i.c. well, IMHO you have two options here:
1158719616 M * Bertl (I'm not getting payed here either :)
1158719647 M * Bertl - have the quota control on the host, update it every now and then and they will work inside the guest
1158719677 M * Bertl - create a new partition (or loopback if you have to) and put that into the guest (or the guest on it)
1158719686 M * Radiance hmm so a user of the guest can be limited by the quota on the host ???
1158719693 M * Bertl yes
1158719703 M * Bertl uids are the same, unless you use xid tagging
1158719706 M * Radiance that's a nice option too
1158719713 M * Radiance nah no xid tagging
1158719746 M * Bertl the users inside might even be able to read their quota values
1158719766 M * Bertl but quota requests/reports have to be done on the host then
1158719807 J * Johnnie ~jdlewis@jdlewis.org
1158719815 M * Bertl wb Johnnie!
1158719818 M * Radiance i think for this specific case i'll create a partition for the vps and move it to that
1158719853 M * Bertl IMHO the better choice anyways .. as it will save you from isues on /var :)
1158719866 M * Radiance true heh
1158719882 M * Radiance thanks mate, appreciate the help
1158719891 M * Bertl you're welcome!
1158720019 M * Radiance btw, you work or still a student ? :)
1158720032 M * ray6 Bertl: isn't the appreciation of lucky users payment enough :)
1158720042 M * Bertl ray6: always :)
1158720065 M * Bertl ray6: but unfortunately that doesn't pay my bills :)
1158720069 M * Radiance probably not, but are you in need of hosting or specific online services ?
1158720072 M * ray6 Bertl: I mean it's you who didn't sell himself to the russians^Wamericans when they asked :)
1158720086 M * Bertl LOL
1158720130 M * Bertl Radiance: /m me some contact information, I'll store it, and when I need something, I'll contact you, okay?
1158720292 Q * Johnnie Quit: G'bye!
1158720389 J * _node ~node@c-69-143-148-254.hsd1.md.comcast.net
1158720405 M * Bertl welcome _node!
1158720610 P * anonc adios
1158720650 M * _node hello Bertl
1158721003 J * Johnnie ~jdlewis@jdlewis.org
1158721802 J * anonc ~anonc@staffnet.internode.com.au
1158721847 M * Bertl wb Johnnie! anonc!
1158721897 M * anonc hola bertl
1158722188 Q * Johnnie Quit: G'bye!
1158722194 J * Johnnie ~jdlewis@jdlewis.org
1158722210 M * Skram Hi, Bertl 
1158722259 M * Johnnie I think my router is done being a horse's patoot!
1158722281 M * Bertl hey Skram! how's going?
1158722334 M * Skram Bertl: *sigh* I am still alive.
1158722344 M * Skram I'll try not to be too dramatic, heh
1158722493 M * Bertl sounds already pretty dramatic to me :)
1158722573 Q * wenchien Ping timeout: 480 seconds
1158722614 M * ray6 alive doesn't sound very dramatic. If he'd say he's dead, that would concern me...
1158722637 M * Johnnie haha
1158722655 Q * yarihm Quit: Leaving
1158722691 M * Skram ray6: we met at HOPE, correct?
1158722696 M * Skram What a riot... haha
1158723012 M * Johnnie Oh, Bertl, OpenVZ bored me to death.
1158723018 M * ray6 Skram: yeah I think so (that we met)
1158723044 M * ray6 after your vserver presentation you mean?
1158723065 M * Skram Uh huh
1158723109 M * Bertl Johnnie: yeah? how so?
1158723129 M * Johnnie Well, sharing timing sources for Asterisk on OpenVZ seems hardly possible in some ways.
1158723148 M * Johnnie I went through all of the moves and did my homework on the device nodes and crap...still no go.
1158723159 M * Bertl ah, what was the problem there?
1158723168 M * Johnnie I do like the kind of control you can accomplish on a network device, but that's about it.
1158723184 M * Bertl the virtual network stack, you mean?
1158723209 M * Johnnie Well, essentially, mknod pretty much does nothing for me... I tried doing stuff on the DEVICES line in the configuration and that was of no help.
1158723214 M * Johnnie Asterisk basically scoffed.
1158723229 M * Johnnie Yes, the virtual network stack...sort of neat as far as iptables go, and what not.
1158723251 M * Johnnie But, I plan on making up for those deficiencies.
1158723266 M * Bertl how so?
1158723289 M * Johnnie I plan on doing a traffic aggregator to firewall and corral traffic to each VPS...external of the VPS host.
1158723305 M * ray6 Skram: or do I mix you up with someone?
1158723314 M * Johnnie It'll also serve other roles, doing rsync backups, serve DNS, miscellaneous routing, et al.
1158723330 M * Bertl Johnnie: yeah, some companies even added some kind of 'external' iptables and routing for the guests
1158723353 M * Johnnie Yeah, it won't be a problem... I need to get Zabbix or Nagios going anyway.
1158723354 M * Bertl basically it is simple to do all this via e.g. a web interface
1158723368 M * Johnnie So, I really need an external system to do that kind of baloney anyhow.
1158723398 M * Johnnie I could be wrong, but from what I read anyway, a user could put the virtual interface in a promiscuous mode anyway.
1158723409 M * Johnnie And I'm not quite keen on that... major security issue.
1158723423 M * Bertl we are also investigating methods to relay iptable and routing commands (addressed to the kernel) to userspace
1158723442 M * Johnnie That could be flaky.
1158723450 M * ray6 if someone wants an own routing table inside a VPS he should use xen :)
1158723452 M * Johnnie Especially if a user does something totally wrong.
1158723456 M * Bertl (might allow to have per guest routing and iptables with proper policy in userspace)
1158723471 M * Johnnie ray6: I'm not that concerned... I'm comfortable with VServer at this point.
1158723508 M * ray6 How far does VZs network virtualization go? They just have virtual interfaces not routing tables do they?
1158723532 M * Bertl layer 2 and 3 actually (in the latest version, IIRC)
1158723533 M * Johnnie You can do routing/firewalling with iptables...if that's what you mean.
1158723545 M * ray6 I mean you would have to add isolation to all the routing stuff and so? Or just give one routing table to a guests IF and reduce control to that one?
1158723545 M * Johnnie Bertl said it better.
1158723546 M * Johnnie haha
1158723563 M * Johnnie ray6: I'm not sure where it places isolation, if any.
1158723577 M * Johnnie OpenVZ even lets you generate MAC addresses per VE.
1158723590 M * Johnnie Nice idea, but another security hole.
1158723592 M * Bertl isolation happens on the (virtual) host bridge if at all
1158723615 M * ray6 Johnnie: that was the question, can I do routing in openVZ? for example start a pppd over ssh and route something into it?
1158723626 M * Bertl yes
1158723628 M * Johnnie ray6: I'm pretty sure, yes.
1158723645 M * Johnnie I haven't done it, but it appears to be fully supported.
1158723672 M * Bertl should be, although I do not know if the ppp interface is explicitely supported
1158723674 M * Johnnie I presume Bertl has tested OpenVZ...I'm sure he's knowledgeable about that.
1158723691 M * Bertl Johnnie: not much testing yet, more code reading :)
1158723695 M * Johnnie Well, PPP shouldn't be a problem...if the kernel is granted support, it should work.
1158723722 M * Johnnie You can definitely do VPN stuff, that part is easy as far as setting up tun/tap and mapping it across VEs.
1158723758 M * ray6 too bad my OpenVZ+Xen kernel can't network at all %-) Will have to try OpenVZ native I guess :)
1158723766 M * Skram :)
1158723767 M * Bertl LOL
1158723790 M * Bertl ray6: still no go? IIRC last time the OVZ folks promised you to fix that, no?
1158723824 M * ray6 Bertl: I haven't had a new try since then... 
1158723833 M * Johnnie I do have to say, with capabilities and flags, you can do a lot of stuff on VServer that you can't do on VZ.
1158723852 M * ray6 Bertl: it's not my main priority as I'm not really planning to use OpenVZ for something
1158723912 M * Bertl i.c.
1158724113 M * ray6 Bertl: my current setup of choice is running Xen on neat large servers to give VMs to projects/customers, and use Vserver inside a xen server to have apache/mysql/sendmail isolated against each other... but it's always good to know what the others can do. Now that FreeBSD+Xen is getting closer I'll also have to look at jails again, but as fas as I can see they're way behind vserver.
1158724160 M * Bertl yeah, for whatever reason they stopped at some point
1158724172 M * ray6 But I knoe people who prefer FreeBSD. But as jail is a normal part of FreeBSD it of course also works inside Xen-FreeBSD
1158724319 M * ray6 Bertl: I think jail was mainly developed by the company who did the initial release and didn't undergo much change since then
1158724346 M * Bertl could be
1158724696 J * cehteh ~ct@pipapo.org
1158724703 M * Bertl wb cehteh!
1158724771 M * ray6 hm, Skram got asleep or doesn't want to be remembered of HOPE? :)
1158724969 M * Skram sorry- Skram is busy with something.. ill be back soon
1158726521 J * KLFMANiK ~mann@people.uvt.tuke.sk
1158726573 M * Bertl welcome KLFMANiK!
1158726610 M * KLFMANiK thanks 
1158726628 M * KLFMANiK i have one question about bind9 in vserver
1158726729 M * KLFMANiK i have kernel 2.6.17.11+vs2.0.2 ... and now bind9 works in vservers without disabling linux-caps ... standard debian packages works
1158726770 M * Bertl hmm, 2.0.2 sounds strange, 2.1.1-rc* would be expected
1158726838 M * Bertl is it a mainline kernel or a debian/distro kernel?
1158726865 M * KLFMANiK it's vanilla + vserver patch ... my recompiled kernel
1158726911 J * gerrit_ ~gerrit@c-67-160-146-170.hsd1.or.comcast.net
1158726942 M * Bertl KLFMANiK: strange, I would opt for a --disable-linux-caps compiled bind
1158726970 M * Bertl (or maybe debian folks added a runtime check or fixed the buggy code?)
1158726980 M * KLFMANiK i recompiled new bind9 packages with disable-linux-caps and then after install i got error: Starting domain name service...:named: -u with Linux threads not supported: no capabilities support or capabilities disabled at build time
1158727030 A * cehteh still opts for a LD_PRELOAD lib which fixes broken apps
1158727039 M * KLFMANiK i check this today more thoroughly ...
1158727070 M * Bertl yeah, please do so, note: 2.1.1 branch allows bindX with linux caps to run as expected
1158727088 M * daniel_hozac KLFMANiK: it means what it says, that's why you should use the patch if you want to use all the BIND features.
1158727133 M * KLFMANiK so my another question ... we have new 2.6.18 kernel ... how you support stale vserver patch? maybe last and last-1 can be supported - like vanilla
1158727137 M * cehteh well .. my idea might be less valuable by time when the vserver kernel handles that :P
1158727140 M * KLFMANiK stale = stable
1158727159 M * Bertl heya 2.6.18 is out!!
1158727183 M * daniel_hozac indeed...
1158727200 M * cehteh doh .. i build a .17 few hours ago
1158727215 M * Bertl KLFMANiK: you mean keeping 2.6.17.x versions (for x >13) or so?
1158727231 M * KLFMANiK yes
1158727235 Q * comfrey Ping timeout: 480 seconds
1158727241 M * KLFMANiK but only for 2.0.2.x
1158727282 M * Bertl KLFMANiK: yeah, well, I think that should be easy, not many changes will be done in 2.0.2 .. only security fixes and really simple cleanups
1158727312 M * KLFMANiK ok ... great ;-)
1158727322 M * Bertl so they should apply to later versions too, and we can always provide an 'updated' release if something breaks
1158727501 M * KLFMANiK yes ... i still want some 'stable' kernel ... but i have some problems with 2.6.17.x because of xfs bug ... so i decided to use last-1
1158727555 M * KLFMANiK now 2.6.18 is 'stable', so no changes expected in 2.6.17.x ... only security and serious bugs
1158727629 M * KLFMANiK so i want vs2.0.2 for 2.6.17.x until 2.6.19 is out ... still last-1 (and maybe vs2.0.2 for last too) .. and of course 2.1.x for last
1158727679 M * KLFMANiK i have another question ... can you provide some 'latest-is' symlink for vserver patch? i'm using LATEST-IS for downloading latest vanilla kernel ...
1158727794 M * Bertl hmm, guess we can do that in the future (but not right now)
1158727821 M * KLFMANiK ok ... in the future - you mean latest-is ?
1158727827 M * Bertl i.e. once we moved to git (will take a week or two) we can do something like that
1158727878 M * KLFMANiK ok ... this can be or not ... 
1158727902 M * Bertl I wonder why the 2.6.18 release was so quick now ...
1158727908 M * Bertl s/quick/early/
1158727927 M * KLFMANiK but more important for me (and maybe other people) is support for last-1 
1158727945 M * Bertl as I said, for stable I do not see a problem there
1158727955 M * KLFMANiK ok, many thanks ;-)
1158727961 M * daniel_hozac it's not particularly hard to do yourself either.
1158727966 M * Bertl on 2.4 we had stable releases spawning 3 ekrnel versions
1158727977 M * Bertl (and sometimes more)
1158728112 M * KLFMANiK yes, i know ... i used vserver maybe from  2.4.21 ... maybe vs1.2.6 ??? very long time ago ;-)
1158728304 M * daniel_hozac KLFMANiK: what should LATEST-IS point to? stable releases or rcs too?
1158728470 M * KLFMANiK stable ..... but it should be LATEST-STABLE and LATEST-DEVEL or something similar ... 
1158728475 M * daniel_hozac Bertl: http://people.linux-vserver.org/~dhozac/p/uv/experimental/util-vserver-0.30.211-rc1.tar.bz2
1158728493 M * daniel_hozac KLFMANiK: yes, but stable releases or really the latest stable, i.e. rcs too?
1158728547 M * KLFMANiK but stable is sufficient ... like vanilla kernel
1158728594 M * Bertl daniel_hozac: excellent! tx!
1158728630 M * Bertl KLFMANiK: guess we will kind of automate that pretty soon
1158728646 M * Bertl (no guarantees though :)
1158728780 M * KLFMANiK ok ... maybe you can ask other people too ;-) not only me
1158728819 M * cehteh wow ... i have a trick which significatly enhances write-speed on md-raid5/ext3 now ;)
1158728837 M * KLFMANiK Bertl: another question ... is some mirror needed? i can make rsync mirror + http, ftp
1158728967 M * KLFMANiK daniel_hozac: for rcs and devel versions we can make some announcement mailing list or RSS ... 
1158729016 M * daniel_hozac cehteh: do tell :)
1158729053 M * Bertl KLFMANiK: not at the moment, but thanks for the offer .. best /msg me with some contact information, so that we can get back to you when we need something
1158729056 M * cehteh daniel_hozac: data=journal with external journal on raid0
1158729059 M * KLFMANiK daniel_hozac: i have script similar to 'websec' package for monitoring changes on http, ftp ... based on diff between two listings ...
1158729087 M * cehteh (and use the  biggiest possible journal)
1158729469 M * Bertl I just figured that we lost an important page on the wiki transition ... the Linux-VServer Users (not hosting providers)
1158729484 M * Bertl i.e. I will add that (after some cleanup) soon
1158729565 J * coocoon ~coocoon@p54A06C15.dip.t-dialin.net
1158729580 M * coocoon morning
1158729593 M * Bertl good morning!
1158731303 M * Bertl okay, I'm off to bed for now ... back later!
1158731313 M * Bertl have a good one everyone, and cya!
1158731318 N * Bertl Bertl_zZ
1158731626 M * nokoya timer
1158732141 J * AjAx-- hiddenserv@tor.noreply.org
1158732375 Q * AjAx- Ping timeout: 480 seconds
1158732383 N * AjAx-- AjAx-
1158733248 Q * cdrx Ping timeout: 480 seconds
1158733938 Q * AjAx- Remote host closed the connection
1158733945 J * AjAx-- hiddenserv@tor.noreply.org
1158733950 N * AjAx-- AjAx-
1158734634 J * wenchien ~wenchien@59-105-176-11.adsl.static.seed.net.tw
1158735120 J * dna_ ~naucki@223-250-dsl.kielnet.net
1158736136 J * Aiken ~james@tooax7-018.dialup.optusnet.com.au
1158736282 Q * Aiken_ Ping timeout: 480 seconds
1158736590 J * cdrx ~legoater@242.32.96-84.rev.gaoland.net
1158736810 J * meandtheshell ~markus@85-125-231-51.dynamic.xdsl-line.inode.at
1158737712 Q * KLFMANiK Quit: [BX] Connection reset by panasync?
1158738701 J * id23 ~id@p50813E62.dip0.t-ipconnect.de
1158738948 Q * AjAx- Remote host closed the connection
1158739125 J * AjAx-- hiddenserv@tor.noreply.org
1158739131 N * AjAx-- AjAx-
1158739844 Q * coocoon Quit: KVIrc 3.2.0 'Realia'
1158741374 Q * rob-84x^ Ping timeout: 480 seconds
1158741779 Q * shedi Quit: Leaving
1158742809 J * Piet hiddenserv@tor.noreply.org
1158743096 J * robig ~robig@154.250.80.212.static.versanetonline.de
1158743104 M * robig hey
1158743121 M * robig :)
1158743152 M * robig is somebody using the newvserver tool from vserver-debiantools package?
1158743199 J * rob-84x^ rob@submarine.ath.cx
1158743732 J * pisco hiddenserv@tor.noreply.org
1158743749 J * derjohn ~derjohn@80.69.37.19
1158743757 M * pisco moin folks
1158743792 M * pisco someone experienced building a fedora guest under  debian vserver host?
1158743919 M * pisco i installed rpmstrap an build a guest via -m rpm switch, but the thes won't build. I think there is an configuration issue ...
1158744949 Q * ruskie Quit: Disconnecting from stoned server.
1158745033 M * waldi okay, 2.6.18 is released, now, wheres the patch? ;)
1158745095 J * ruskie ~ruskie@ruskie.user.oftc.net
1158745177 M * derjohn waldi, good Q! 
1158745196 M * derjohn waldi, doesnt rc35 apply clean? (I didnt try yet)
1158745217 J * lilalinux ~plasma@dslb-084-058-204-047.pools.arcor-ip.net
1158745408 J * pisc1 hiddenserv@tor.noreply.org
1158745435 Q * pisco Ping timeout: 480 seconds
1158745443 M * robig I got a problem: * File '/proc/sys/kernel/vshelper' does not exist but is required for vshelper setup
1158745444 M * robig what to do? is the kernel prepared correctly? testme.sh says averythis is ok.
1158745597 M * derjohn doener, daniel_hozac : did anyone do already 2.6.18 work ?
1158746401 Q * cdrx Ping timeout: 480 seconds
1158747166 J * AjAx-- hiddenserv@tor.noreply.org
1158747440 Q * AjAx- Ping timeout: 480 seconds
1158747444 N * AjAx-- AjAx-
1158747763 P * pisc1 
1158748660 J * fluor ~fluor@tanneries.squat.net
1158748763 J * joep ~Joep@c53753253.cable.wanadoo.nl
1158748789 P * joep 
1158748806 M * nayco hello, all !
1158749203 Q * Aiken Quit: Leaving
1158749328 J * kaner_ kaner@strace.org
1158749376 Q * kaner Remote host closed the connection
1158749664 M * robig hello. can anyone help me with this: * File '/proc/sys/kernel/vshelper' does not exist but is required for vshelper setup
1158749668 M * robig ?
1158749745 M * mnemoc load the right kernel
1158749968 J * pisco hiddenserv@tor.noreply.org
1158749984 J * cdrx ~legoater@cap31-3-82-227-199-249.fbx.proxad.net
1158750269 Q * pisco Remote host closed the connection
1158750561 M * robig testme.sh says that averythis is ok
1158751761 M * robig I have no /proc/sys directory... 
1158751804 M * robig I think I've to enable sysvfs support?
1158751894 Q * morfoh Quit: Lost terminal
1158752060 Q * AjAx- Ping timeout: 480 seconds
1158752420 J * shedi ~siggi@dsl-149-109-85.hive.is
1158753459 M * weasel hmm
1158753583 M * weasel ylvi:~# vserver vtest start
1158753583 M * weasel vshelper.init: can not determine xid of vserver 'vtest'; returned value was ''
1158753605 M * weasel according to google I'm not the only one seeing this, however none of the first few threads had a solution.
1158753608 M * weasel any ideas?
1158753917 M * weasel ah
1158753924 M * weasel no processes running.
1158753930 M * weasel there should be a start and enter then
1158754858 M * harry how did you create your vserver?
1158754869 M * harry or: what's in /etc/vservers/vtest/context ?
1158755108 M * weasel vserver build ... debootstrap
1158755116 M * weasel whatever I put in there.  say 51.
1158755137 M * weasel the problem was/is that you can't start a vserver unless you have daemons that start with init.
1158755509 M * harry btw. Bertl_zZ etc... 2.6.18 has been released
1158755859 M * robig the patch for 2.6.17.13 is stable.. or?
1158756161 Q * marl Read error: Connection reset by peer
1158756451 M * daniel_hozac robig: yes.
1158756653 Q * _node Ping timeout: 480 seconds
1158756993 Q * Piet Read error: Connection reset by peer
1158757033 J * AjAx-- hiddenserv@tor.noreply.org
1158757055 N * AjAx-- AjAx-
1158757058 J * Piet hiddenserv@tor.noreply.org
1158757214 M * robig daniel_hozac: can I use any util-vserver with it?
1158757478 M * harry daniel_hozac: ever heared anything from enrico?
1158757691 A * harry looking for an interdiff for 2.0.2 to 2.0.2.1
1158757704 M * harry btw. what's the correct way: 2.02 or 2.0.2
1158757720 M * harry because the patch 2.0.2.1, and 2.02
1158757725 M * harry so what's the deal here?
1158757987 Q * nebuchadnezzar Read error: Connection reset by peer
1158758473 J * nebuchadnezzar ~nebu@zion.asgardr.info
1158758857 M * |yang| hey doener
1158759081 Q * AjAx- Remote host closed the connection
1158759351 J * coocoon ~coocoon@p54A06616.dip.t-dialin.net
1158759405 M * robig hm.. I still get the error: '/proc/sys/kernel/vshelper' does not exist but is required for vshelper setup
1158759474 M * harry did you do a make install-distribution 
1158759474 M * harry ?
1158759486 M * harry in the util-vserver sourcedir?
1158759505 J * vasko ~vasko@unreal.rainside.sk
1158759570 M * robig yes
1158759592 M * robig I don't have a sys direcory in /proc !
1158759628 M * robig so any idea why?
1158759717 J * Borg- ~borg@217.97.139.162
1158760068 M * doener robig: forgot to enable CONFIG_SYSCTL?
1158760078 M * doener morning folks
1158760094 Q * Zaki reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * vasko reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * kaner_ reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * transacid reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * lilalinux reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * derjohn reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * robig reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * id23 reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * gerrit_ reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * Johnnie reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * anonc reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * weeble reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * mugwump reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * Loki|muh reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * node reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * ntrs reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * Borg- reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * Piet reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * micah reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * eyck reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * ray6 reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * phreak`` reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * MooingLemur reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * nayco reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * starlein reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * tanjix reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * FireEgl reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * fosco reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * weasel reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * Hunger reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * coocoon reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * nebuchadnezzar reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * mnemoc reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * litage reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * phedny reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * mountie reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * sladen reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * Greek0 reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * FloodServ reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * samuel_ reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * matti reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * |yang| reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * nokoya reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * mcp reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * cryptronic reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * ebiederm reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * [PUPPETS]Gonzo reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * click reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * bogus reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * tokkee reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * s0undt3ch reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * michal` reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * tso reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * harry reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * ruskie reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * Snow-Man reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * ag- reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * DreamerC reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * duckx reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * ay_ reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * Wonka reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * cohan_ reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * daniel_hozac reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * nib-nico reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * nox reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * dhansen reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * shedi reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * cdrx reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * rob-84x^ reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * dna_ reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * gdm reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * Vudumen reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * MrX reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * fs reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * ex reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * Medivh reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * bragon reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * meebey reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * SNy reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * AndrewLee reticulum.oftc.net hydrogen.oftc.net
1158760094 Q * pusling reticulum.oftc.net hydrogen.oftc.net
1158760125 J * transacid ~transacid@transacid.de
1158760125 J * kaner_ kaner@strace.org
1158760125 J * vasko ~vasko@unreal.rainside.sk
1158760125 J * the_hydra ~a_mulyadi@202.59.168.29
1158760125 J * Borg- ~borg@217.97.139.162
1158760125 J * coocoon ~coocoon@p54A06616.dip.t-dialin.net
1158760125 J * nebuchadnezzar ~nebu@zion.asgardr.info
1158760125 J * Piet hiddenserv@tor.noreply.org
1158760125 J * lilalinux ~plasma@dslb-084-058-204-047.pools.arcor-ip.net
1158760125 J * derjohn ~derjohn@80.69.37.19
1158760125 J * robig ~robig@154.250.80.212.static.versanetonline.de
1158760125 J * id23 ~id@p50813E62.dip0.t-ipconnect.de
1158760125 J * gerrit_ ~gerrit@c-67-160-146-170.hsd1.or.comcast.net
1158760125 J * Johnnie ~jdlewis@jdlewis.org
1158760125 J * anonc ~anonc@staffnet.internode.com.au
1158760125 J * FireEgl FireEgl@Sebastian.Atlantica.US
1158760125 J * samuel_ ~samuel@jupe.quebectelephone.com
1158760125 J * matti matti@linux.gentoo.pl
1158760125 J * |yang| ~yang@cpe-213-157-253-172.dynamic.amis.net
1158760125 J * nokoya young@hi-230-82.tm.net.org.my
1158760125 J * mcp ~hightower@wolk-project.de
1158760125 J * cryptronic crypt@mail.openvcp.org
1158760125 J * fosco fosco@konoha.devnullteam.org
1158760125 J * weasel ~weasel@weasel.noc.oftc.net
1158760125 J * micah ~micah@micah.riseup.net
1158760125 J * s0undt3ch ~s0undt3ch@bl7-245-194.dsl.telepac.pt
1158760125 J * bogus ~bogusano@fengor.net
1158760125 J * tokkee tokkee@casella.verplant.org
1158760125 J * FloodServ services@services.oftc.net
1158760125 J * harry ~harry@d54C2508C.access.telenet.be
1158760125 J * tso ~tso@196-019.adsl.pool.ew.hu
1158760125 J * ntrs ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com
1158760125 J * Greek0 ~greek0@85.255.145.201
1158760125 J * node ~dwindsor@stanford.columbia.tresys.com
1158760125 J * click click@ti511110a080-5865.bb.online.no
1158760125 J * Hunger Hunger.hu@Hunger.hu
1158760125 J * nayco ~nayco@proxy2.laroche.univ-nantes.fr
1158760125 J * starlein star@fo0bar.de
1158760125 J * MooingLemur ~troy@shells200.pinchaser.com
1158760125 J * tanjix ~tanjix@office.star-hosting.de
1158760125 J * phreak`` ~phreak``@140.211.166.183
1158760125 J * [PUPPETS]Gonzo gonzo@langweiligneutral.deswahnsinns.de
1158760125 J * michal` ~michal@www.rsbac.org
1158760125 J * Loki|muh loki@satanix.de
1158760125 J * sladen paul@starsky.19inch.net
1158760125 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com
1158760125 J * phedny ~mark@volcano.p-bierman.nl
1158760125 J * litage ~nick@203.220.55.70
1158760125 J * mnemoc ~amery@kilo105.server4you.de
1158760125 J * ray6 ~ray@194.126.159.45
1158760125 J * ebiederm ~eric@ebiederm.dsl.xmission.com
1158760125 J * eyck eyck@195.242.124.71
1158760125 J * mugwump ~samv@watts.utsl.gen.nz
1158760125 J * weeble ~weeble@81.52.144.1
1158760139 J * ruskie ~ruskie@ruskie.user.oftc.net
1158760139 J * Snow-Man ~sfrost@kenobi.snowman.net
1158760139 J * ag- ~ag@caladan.roxor.cx
1158760139 J * DreamerC ~dreamerc@59-115-49-95.dynamic.hinet.net
1158760139 J * duckx ~Duck@tox.dyndns.org
1158760139 J * ay_ ay@false.linpro.no
1158760139 J * Wonka produziert@chaos.in-kiel.de
1158760139 J * nib-nico ~nico@nibweb.net
1158760139 J * cohan_ ~cohan@koniczek.de
1158760139 J * nox ~nox@nox.user.oftc.net
1158760139 J * dhansen ~dave@sprucegoose.sr71.net
1158760139 J * daniel_hozac ~daniel@c-2c1472d5.010-230-73746f22.cust.bredbandsbolaget.se
1158760152 J * Zaki ~Zaki@88.213.22.103
1158760165 J * shedi ~siggi@dsl-149-109-85.hive.is
1158760165 J * cdrx ~legoater@cap31-3-82-227-199-249.fbx.proxad.net
1158760165 J * rob-84x^ rob@submarine.ath.cx
1158760165 J * dna_ ~naucki@223-250-dsl.kielnet.net
1158760165 J * gdm ~gdm@www.iteration.org
1158760165 J * Vudumen ~vudumen@perverz.hu
1158760165 J * MrX ~urk@219.95.2.216
1158760165 J * fs fs@213.178.77.98
1158760165 J * AndrewLee ~andrew@tnlug.linux.org.tw
1158760165 J * ex ex@valis.net.pl
1158760165 J * meebey meebey@booster.qnetp.net
1158760165 J * pusling pusling@195.215.29.124
1158760165 J * bragon ~weechat@sd866.sivit.org
1158760165 J * SNy 6cfbac777d@bmx-chemnitz.de
1158760165 J * Medivh ck@paradise.by.the.dashboardlight.de
1158760315 M * harry robig: do you have proc enabled in your kernel?
1158760327 M * harry do you have proc mounted ? is it in fstab?
1158760490 M * robig proc is available, but there ist no sys dir in it
1158760638 M * doener 15:47:48 <doener> robig: forgot to enable CONFIG_SYSCTL?
1158760646 M * derjohn weasel, it that also true if you set initstyle "plain"? 
1158760654 M * doener (just to make sure that it wasn't lost in the netsplit)
1158760669 M * derjohn doener, hey! did you try to patch 2.6.18 already?
1158760692 M * doener oh, it's out now?
1158760697 M * doener I just got up ;)
1158760707 M * robig # grep SYSCTL .config
1158760707 M * robig CONFIG_SYSCTL=y 
1158760721 J * AjAx-- hiddenserv@tor.noreply.org
1158760721 M * derjohn doener, yes, since 0400 UTC ;)
1158760727 N * AjAx-- AjAx-
1158760734 M * doener derjohn: no, 0357 UTC ;)
1158760738 M * robig have to go.. :( see you later.. 
1158760744 P * robig 
1158760766 M * doener I started a port to -rc6 some time ago, but got stuck
1158760783 M * derjohn doener, and it (2.6.17.13) fails on many hunks. I didnt try fuzz yet, just wanted to know if anyone worked on 2.6.18-rcs ...
1158760791 M * derjohn doener: pointy haired ;)
1158760865 M * doener derjohn: most failed hunks are trivial, I got lost in ebiederm's proc changes
1158760950 M * ebiederm Hmm.  That is unfortunate.
1158760958 M * derjohn ebiederm, is a vserver-saboteur ;)
1158761063 M * ebiederm If someone wants to work on it i can answer questions.   So far I haven't done anything except work on making many small changes to clean up /proc
1158761067 M * doener nah, I'm just not confident enough about what that part of the code (each, vanilla and vserver) really does, and didn't have the time to find out
1158761201 Q * rob-84x^ Read error: Connection reset by peer
1158761286 Q * the_hydra Read error: Connection reset by peer
1158761359 Q * AjAx- Ping timeout: 480 seconds
1158761390 M * ebiederm Ugh.  I just realized that isn't the -mm tree you are talking about!
1158761429 M * ebiederm As a wishlist for some brave person porting vserver to the -mm tree so the work on starting to use the new stuff the kernel provides would be nice.
1158761469 M * doener ebiederm: I intended to give it a try after the .18 port, but the latest -mm announcement was scary
1158761485 M * ebiederm Yeah I guess so.
1158761642 J * AjAx-- hiddenserv@tor.noreply.org
1158761648 N * AjAx-- AjAx-
1158761721 Q * AjAx- 
1158761784 J * pisc1 hiddenserv@tor.noreply.org
1158761885 M * mnemoc doener: can you summarize it please? :)
1158762029 M * doener mnemoc: hm?
1158762053 M * mnemoc "but the latest -mm announcement was scary"
1158762070 M * mnemoc three words summary? :) please
1158762111 M * mnemoc or the date of the mail :\ i can't find it :(
1158762177 M * ebiederm Subect: 2.6.18-rc7-mm1
1158762185 M * doener three words: Andrew said so
1158762250 M * doener search for "dogpile" in that mail ;)
1158762356 M * mnemoc uh
1158762427 M * meandtheshell derjohn: I fixed some typos etc. within the FAQ@new_wiki - by the way I have another question - shoudn't we make a s/master/host/ here http://linux-vserver.org/Frequently_Asked_Questions#Trying_to_connect_to_a_vserver_from_the_host_or_another_vserver_on_the_same_host_fails
1158762438 M * ebiederm Date: Tue, 19 Sep 2006 02:28:48 -0600
1158762464 M * ebiederm The quote from andrew: - It took maybe ten hours solid work to get this dogpile vaguely
1158762464 M * ebiederm   compiling and limping to a login prompt on x86, x86_64 and powerpc. 
1158762464 M * ebiederm   I guess it's worth briefly testing if you're keen.
1158762468 M * derjohn meandtheshell, thankx !
1158762505 M * meandtheshell derjohn: no problem - but am I right - dont't know - maybe you had something special in mind with the term master - no?
1158762596 M * derjohn meandtheshell, that FAQ part isnt written by /me. My task is just merging and structuring the stuff.
1158762620 M * derjohn meandtheshell, correntions welcome. I wull not proofread before restructuring
1158762622 M * meandtheshell derjohn: you're right - my fault :)
1158762625 M * derjohn *will
1158762659 M * derjohn meandtheshell, hm? nono very weclome! But I cannot tell you what the original author meant with "master" ...
1158762691 N * Bertl_zZ Bertl
1158762695 M * Bertl morning folks!
1158762706 M * derjohn Bertl, Hi!
1158762706 M * meandtheshell derjohn: same here - I'd suggest to make a s/master/host/ - what do you think?
1158762717 M * meandtheshell Bertl: hi there :)
1158762722 M * derjohn Bertl, saw the 2.6.18 realease ?
1158762725 M * meandtheshell morning Vienna :)
1158762731 M * derjohn meandtheshell, do it ! ;)
1158762735 M * Bertl derjohn: yep, right before I went to bed ...
1158762738 M * meandtheshell derjohn: ok
1158762763 M * Bertl seems to be a huge pile of changes ...
1158762790 M * Bertl but it seems that 2.6.19 will contain even more changes
1158762815 M * derjohn Bertl, is that a good sign a preview of horror?
1158762831 M * Bertl only time can tell :)
1158762841 M * derjohn I still press thumbs that the sata and md stuff gets fully working. 
1158762888 M * derjohn (last time i wanted to install a md on nforce with a 2.6.17 based installer. doesnt work, gives 0x50 errors. No errors without md)
1158762903 M * derjohn 2.6.15 worked on the same machine.
1158763391 A * doener goes porting 2.0.2 to 2.6.18
1158763404 M * doener ehrm, 2.0.2.1 or whatever
1158763433 A * derjohn tries to convince doener to "devel" :9
1158763559 A * doener resists
1158763724 J * comfrey ~comfrey@h-64-105-215-75.sttnwaho.covad.net
1158763823 M * Bertl welcome comfrey!
1158763843 M * SNy doener: "Do, or do not. There is no 'try'." --Yoda
1158763850 M * SNy Uh, derjohn, even.
1158763852 M * Bertl :)
1158763863 J * Roey ~katz@h-69-3-4-130.mclnva23.covad.net
1158763869 M * Bertl wb Roey!
1158763886 M * harry is there a code difference in 2.0.2 and 2.0.2.1 ?
1158763891 M * doener SNy: I _do_ resist ;)
1158763891 M * harry or are they exactly the same?
1158763909 M * SNy doener: Yah, wrong nick completion.
1158763915 Q * Roey 
1158763931 J * Roey ~katz@h-69-3-4-130.mclnva23.covad.net
1158764272 M * derjohn harry, they differnt slightly. dont take 2.0.2.
1158764382 M * |yang| oh a nice new website !
1158764528 Q * FireEgl Read error: Connection reset by peer
1158764660 Q * coocoon Quit: KVIrc 3.2.0 'Realia'
1158764824 J * Hollow_mobile ~bene@217.110.45.98
1158764827 A * Hollow_mobile waves
1158765000 J * stefani ~stefani@tsipoor.banerian.org
1158765002 J * rob-84x^ rob@submarine.ath.cx
1158765132 J * coocoon ~coocoon@p54A078F6.dip.t-dialin.net
1158765287 M * Bertl hey Hollow_mobile!
1158765303 M * Bertl wb stefani! rob-84x^! wb coocoon !
1158765317 M * stefani hola.
1158765327 M * Bertl Hollow_mobile: is the old wiki history completely borked? (i.e. do you know?)
1158765519 Q * pisc1 Ping timeout: 480 seconds
1158765637 Q * Borg- Quit: leaving
1158766058 M * Hollow_mobile Bertl: yeah, it's quite borked... also all pages with umlauts or similar characters are borked...
1158766152 M * Bertl okay, np
1158766180 M * coocoon Hollow_mobile: hello hollow, please tell me short which package manager vcd supports for templates
1158766229 M * coocoon +shortly
1158766322 J * harti ~hw@83-215-237-5.seek.stat.salzburg-online.at
1158766367 M * Hollow_mobile coocoon: none currently
1158766412 M * coocoon Hollow_mobile: cool heh, but skeleton is availabe?
1158766418 M * Hollow_mobile but i'm off again, will come back the next days...
1158766428 M * coocoon ok 
1158766495 M * daniel_hozac coocoon: it uses tarballs only, AFAIK.
1158766512 M * Hollow_mobile cu soon
1158766519 M * coocoon daniel_hozac: ok will see thanx
1158766519 M * daniel_hozac Hollow_mobile: have fun!
1158766533 Q * Hollow_mobile Write error: connection closed
1158766570 M * daniel_hozac harry: http://people.linux-vserver.org/~dhozac/p/uv/experimental/util-vserver-0.30.211-rc1.tar.bz2 http://archives.linux-vserver.org/200609/0111.html
1158766905 J * tatiane ~tatiane@201009108033.user.veloxzone.com.br
1158766949 Q * tatiane 
1158767545 Q * harti Quit: Leaving
1158767660 M * nox daniel_hozac: is there a changelog anywhere or will this come?
1158767818 M * daniel_hozac nox: it's in the tarball, as well as the repository phreak`` linked in his mailing list post.
1158768056 Q * comfrey Ping timeout: 480 seconds
1158768186 M * nox oh detailed (;
1158768220 M * Bertl okay, off for dinner now .. back shortly
1158768279 M * daniel_hozac nox: well, it is generated from the commits, so... yeah ;)
1158768699 Q * duckx Remote host closed the connection
1158768861 J * bonbons ~bonbons@83.222.36.111
1158768965 Q * gerrit_ Ping timeout: 480 seconds
1158769020 J * Borg- ~borg@cube.benet.uu3.net
1158769373 Q * meandtheshell Quit: exit (0);
1158769638 M * coocoon daniel_hozac: hello 
1158769653 M * coocoon daniel_hozac: do u know something about whirlpool hash
1158769673 M * Bertl back now ...
1158769679 M * coocoon daniel_hozac: and how to tell my system that it shall use it instead of standard hash
1158769686 M * Bertl coocoon: wikipedia knows :)
1158769695 M * coocoon ah ok
1158769701 M * coocoon will have a look there 
1158769716 Q * fluor Quit: .
1158770045 M * coocoon Bertl: i am asking because of try to use vcd and hollow has implemented whirlpool hash, so now i got everytime authentication errors
1158770070 M * coocoon Bertl: holhas implemented in lucid i think and i have installed the latest lucd
1158770080 M * Bertl implemented or used?
1158770090 M * daniel_hozac used.
1158770097 M * coocoon ah ok 
1158770114 M * coocoon so i just wait till hollow is back
1158770140 M * Bertl yeah, probably best .. try to collect as much info as possible
1158770194 M * daniel_hozac whirlpool replaced the broken hash algorithm that was there previously, because it was segfaulting for mnemoc, IIRC.
1158770232 M * coocoon daniel_hozac: yes right, but how has he used it only for hashing the passwd for vxdb
1158770273 M * daniel_hozac that was my impression.
1158770319 M * coocoon hm, how can i tell my system that it shall has to whirlpool, by sending vcc ... commands
1158770359 M * daniel_hozac i'd assume vcc/vcd does it as necessary.
1158770425 M * coocoon it does not because of the hashs differs , i have installed jacksum where also whirlpool is implemented but maybe i must install whirlpool alone
1158770491 M * daniel_hozac can't you just use vxpasswd?
1158770580 M * coocoon yes i can, will have a look 
1158770616 M * coocoon daniel_hozac: this will hash with which hash function?
1158770624 M * daniel_hozac whirlpool.
1158770667 M * coocoon ah ok maybe it is false in vxdb, ok thanx will have a look
1158770813 M * coocoon daniel_hozac: cool thanx it was false in the vxdb file thanx a lot ;-)
1158770830 M * coocoon daniel_hozac: dunno before for what to use vxpasswd 
1158770836 M * coocoon daniel_hozac: thanx a lot
1158770845 M * daniel_hozac hehe, np.
1158770911 M * doener Bertl: why is the lock xid field copied in locks_copy_lock instead of __locks_copy_lock?
1158771030 J * bazkie ~cosmo@cc896613-a.eelde1.dr.home.nl
1158771039 J * mijke ~mijke@53530101.cable.casema.nl
1158771043 M * bazkie lol
1158771044 M * mijke omg druk
1158771076 P * mijke moiuhhhh
1158771367 J * gerrit_ ~gerrit@bi01p1.co.us.ibm.com
1158771572 M * Bertl hmm, welcome bazkie!
1158771640 M * Roey druk as in drunk?
1158771668 N * Belu_zZz Belu
1158771940 M * bazkie tnx Bertl!
1158771957 M * bazkie gotta fetch me breakfast
1158772088 J * Piet_ hiddenserv@tor.noreply.org
1158772277 M * Roey bazkie:  breakfast, yarrrrrrrrr
1158772308 M * bazkie doef tjik doefe doef tjik!!
1158772314 Q * Piet Ping timeout: 480 seconds
1158772510 M * Roey what's that??
1158772567 M * bazkie a breakbeat!
1158772573 M * bazkie breakfastbeat
1158772578 M * bazkie yarrr
1158772866 Q * shedi Ping timeout: 480 seconds
1158773366 J * Nei ~ailin@userv2.informatik.uni-leipzig.de
1158773424 M * Nei has anyone done vservers in vservers yet, or maybe management scripts that allow a vserver owner to create other vservers (but with fixed restrictions)?
1158773502 J * shedi ~siggi@dsl-149-109-85.hive.is
1158773645 Q * shedi 
1158773886 J * Piet__ hiddenserv@tor.noreply.org
1158774174 Q * Piet_ Ping timeout: 480 seconds
1158774218 M * Bertl Nei: nope, but it is planned in the future ...
1158774236 M * Bertl doener: no idea ... 
1158774335 M * doener damn ;)
1158774636 M * phreak`` bah, I just had a 4+ hours meeting with IBM's storage division :(
1158774658 M * phreak`` heya * by the way (especially doener and Bertl)
1158774702 M * Bertl ah, informative?
1158774742 M * phreak`` pretty much (hah, we know more than we knew before ;), guess thats what you call informative :)
1158774759 M * Nei cool maybe it will exist one day :)
1158774800 M * ray6 nei: easiest would be a script on the host which the guest can activate using an ssh key limited to this specific script I would say
1158774845 M * Nei ray6 but such a script needs to be written first, and in a secure way ;)
1158775125 J * comfrey ~comfrey@h-64-105-215-75.sttnwaho.covad.net
1158775282 J * Blissex ~Blissex@82-69-39-138.dsl.in-addr.zen.co.uk
1158775291 M * Bertl wb comfrey! Blissex!
1158775307 M * Blissex hi!
1158775358 J * wdv ~wouter@wdvmac.student.utwente.nl
1158775430 M * wdv hi.. I am trying to get bind9 running on a debian vserver.. but it doesn't work.. I know about the causes so I am looking for an up to date debian bind9 package which works.. because the one provided on the site doesn't work
1158775446 M * Bertl welcome wdv!
1158775452 M * Nei thats interesting
1158775460 M * wdv thanks
1158775462 M * Bertl wdv: well, basically you have several options
1158775468 M * Nei I havent yet looked into bind but I also read the wiki page describing the problems
1158775485 M * Bertl wdv: use a recent devel kernel (and bind will work out of the box)
1158775503 M * Nei I imagine I will want to run it in a vserver some day too so I'll listen closely ;)
1158775505 M * wdv Bertl, unfortunately I don't administer the box.. I only have an account
1158775505 M * Bertl wdv: (re)compile the bind9 (package) without linux caps
1158775513 M * Nei do you know how recent recent is?
1158775527 M * wdv ?
1158775528 M * Bertl Nei: I'd advise 2.1.1-rc35
1158775568 M * wdv Bertl, how can I recompile the standard debian package without linux caps?
1158775574 M * Bertl wdv: I have no idea about debian packages, but simply adding --disable-linux-caps (or something like that) is sufficient on the ./configure
1158775581 M * Nei ah, so you need a recent vserver patch
1158775593 M * Nei debian packages are pretty nice
1158775605 M * Bertl wdv: the problem is more a bug in bind than anything else
1158775605 M * Nei if you download the package source with apt
1158775609 M * wdv yeah, unfortunately I am not really into debian and I am setting this up for a friend ;)
1158775618 M * Nei just put that configure flag in debian/control
1158775622 M * Bertl wdv: they use the linux caps in a wrong way ...
1158775631 M * Nei and then use dpkg-buildpackage
1158775638 M * Nei thats all you need to do really
1158775642 M * Nei and will end ub with nice .debs
1158775684 M * Nei (maybe tag the version as custom so it is not affected by automatic updates. obviously this has the security drawback, although the official debs wouldnt work anyway I guess ;p)
1158775709 M * wdv Fetched 4656kB in 3s (1227kB/s)
1158775709 M * wdv sh: line 1: dpkg-source: command not found
1158775709 M * wdv Unpack command 'dpkg-source -x bind9_9.2.4-1.dsc' failed.
1158775709 M * wdv E: Child process failed
1158775716 M * wdv and apparently I can't install dpkg-source either
1158775727 M * wdv E: Couldn't find package dpkg-source
1158775753 M * Nei welllll
1158775771 M * Nei obviously the equation binary name == package name does not necessarily hold tru
1158775772 M * Bertl wasn't there some dpkg cache search thingy?
1158775778 M * Nei try aptitude istall dpkg-dev 
1158775779 M * doener wdv: dpkg-dev
1158775780 M * wdv dpkg-dev does the trick :)
1158775819 M * doener and I'd use build-essential anyway ;) (depends on dpkg-dev)
1158775836 M * Nei yea build-essential you should really install that one
1158775849 M * wdv hmm.. I can't find anything about ./configure lines in debian/control
1158775915 M * wdv am I right that I have to look into debian/rules?
1158775965 M * Nei thats because I'm talking bs
1158775970 M * Nei rules of course, sorry
1158775973 M * wdv :):)
1158775990 M * Nei but you found it yourself anyway ;)
1158776016 M * wdv yeah I am not entirely clueless.. I am more into freebsd ;)
1158776033 M * Nei to each his own
1158776040 M * Nei binary packages sure are convenient though
1158776053 M * wdv freebsd has binary packages too ;)
1158776096 M * wdv it's compiling
1158776099 M * wdv thanks so far :)
1158776148 M * Nei so basically its recompiling bind or using the exp version
1158776158 M * Nei not that big a deal it seems
1158776177 M * wdv nah indeed.. I think I can handle it from here
1158776192 J * s0undt3ch_ ~s0undt3ch@82.155.71.48
1158776211 M * Nei I'm just writing it down again for myself too, since I might want to set this up as well
1158776380 M * Nei btw it even had these exact steps outlined on the wiki page :D including the correct file of debian/rules
1158776408 M * wdv ohw :)
1158776413 M * wdv I only looked at the FAQ
1158776440 Q * s0undt3ch Read error: Operation timed out
1158776440 N * s0undt3ch_ s0undt3ch
1158776905 Q * dna_ Read error: Connection reset by peer
1158777003 M * essobi_ FreeBSD is nice.
1158777007 M * essobi_ Umm.
1158777016 M * essobi_ Can you install a Vserver from an Iso?
1158777017 M * essobi_ :)
1158777030 M * essobi_ I've got an Iso I wanna try out..
1158777038 M * essobi_ minimalistic server..
1158777068 M * daniel_hozac you can't run the ISO, but you should be able to mount it and extract it.
1158777076 M * Bertl yes, you can install vservers from an iso, but not freebsd one :)
1158777159 M * essobi_ Bertl:  I assumed as much. ;)
1158777172 M * essobi_ I've got a minimalistic linux Iso I wanna try.. any suggestions?
1158777198 M * essobi_ Or would it just be easier to install it and tar up the entire thing and untar it over a vserver?
1158777314 M * matti Hi Bertl, daniel_hozac :)
1158777337 M * Bertl evening matti!
1158777368 M * Nei when trying to provide virtual web hosting on a single ip, are there any strategies to provide ftp access to customers?
1158777389 M * Nei or is non-standard and different port for each vserver the only way out
1158777396 M * Borg- Nei: sure.. use virtual ftp accounts
1158777422 M * Borg- user@domain login
1158777426 M * Nei supposed each user has its own vserver, I cant quite see that pan out
1158777460 J * bronson ~bronson@66.160.177.212
1158777468 M * Bertl welcome bronson! hey Borg-!
1158777469 M * Nei or is there an ftp that could run on the host and tunnel through to the vservers or something equally evil?
1158777487 M * bronson Hola Bertl.
1158777499 M * Bertl Nei: certain services (read: those who do not depend on end to end security) can be mapped/proxied
1158777518 M * Bertl Nei: which allows to map them to guests on virtual ips
1158777520 M * bronson Quick question...  I remember on the old linux-vserver site some docs telling how to restrict a vserver to using a single network interface.
1158777527 M * bronson Can't find them now...  Any ideas?
1158777540 M * Nei any details/tutorials?
1158777557 M * Bertl bronson: I think you remember wrong, Linux-VServer is not interface related, it is ip based only
1158777592 M * Bertl bronson: i.e. if you assign a certain ip, the interface it is assigned to will be available ... not the other way round
1158777594 J * dna_ ~naucki@223-250-dsl.kielnet.net
1158777608 M * Bertl wb dna_!
1158777631 M * bronson Could be.
1158777642 M * bronson Restricting a vserver to a single ip is fine too.
1158777655 M * bronson Bertl, do you know where docs on this might be?
1158777663 M * bronson Erm...  what about loopback?
1158777663 M * Bertl that is what you do when you assign a single ip to the guest 
1158777690 M * Bertl bronson: loopback is usually not part of the guest
1158777708 M * Bertl bronson: btw, you an find the old wiki under oldwiki.linux-vserver.org
1158777747 M * Bertl bronson: just using --interface ... on guest creation will restrict the guest to those ips
1158777748 M * bronson So, if I restrict a vserver to a single ip, and ssh listens on all interfaces, it ends up only listening on the IP that I specified?
1158777755 M * bronson Cool.
1158777757 M * Bertl precisely
1158777768 M * bronson That should do...  But I'm a little worried about lack of loopback.
1158777770 M * Bertl only on the host you have to take care of such services yourself
1158777778 M * Borg- fuck
1158777783 M * Borg- so there is no loopback?
1158777792 M * daniel_hozac not yet.
1158777801 M * Bertl bronson: loopback is usually handled by remapping the localhost to the first assigned ip
1158777802 M * Nei what do you need to lo for anyway
1158777803 M * bronson Bertl, no problem, I won't be running any listeners on the host.
1158777810 M * Borg- dammit ;/ I could be fried then.
1158777817 Q * Zaki charon.oftc.net europa.oftc.net
1158777829 M * Bertl Borg-: works with basically all services which are not broken  by design :)
1158777849 M * Nei Bertl so how do i map/proxy ftp?
1158777863 M * ray6 bertl: when I use ssh-forwarding without -g it usually binds just 127.0.0.1 - that won't work then?
1158777881 M * Bertl Borg-: nobody says that localhost must be 127.0.0.1, neither does any assumption hold that those ips are on lo
1158777892 J * Zaki ~Zaki@88.213.22.103
1158777905 M * Bertl ray6: if it does so, it is broken, if it looks up localhost, then it will work fine :)
1158777977 M * Borg- Bertl: I know.. but I have quite specic needs here
1158777991 M * ray6 Borg: which are...?
1158777996 M * Bertl Borg-: well, you -can_ assign 127.0.0.1 to the guest too
1158777998 M * Borg- Bertl: you are wrong.. 127,0.0.0/8 is loopback.. and nothing else
1158778019 M * Bertl Borg-: who says so? :)
1158778022 A * matti hands a cup of coffee to Bertl :)
1158778031 M * Borg- ray6: im trying to prepare test server.. so few developers can test some network proggies
1158778031 M * ray6 Bertl: ssh looks up localhost for that, OK
1158778051 M * Borg- Bertl: RFC
1158778073 M * Bertl Borg-: now you are talking about the network addresses, not lo, right?
1158778089 M * Borg- Bertl: isnt it related? as we are talking about IPv4
1158778105 M * Bertl well, okay, so you need a loopback address, that's it?
1158778119 M * Bertl (or in contrast, you want to have 'lo' :)
1158778120 M * Borg- many loopbacks addresses.. all w/ 127.0.0.1
1158778141 M * Bertl or maybe you just want 'guest local' traffic?
1158778152 M * Bertl those are all different things :)
1158778165 M * Borg- Bertl: okey.. im new to vserver.. so I dont know how exacly its wroking..
1158778182 M * Borg- all I need is to have virtualized linux.. so users can do compile and test they network proggies
1158778191 M * Bertl it's quite simple, think like 'ip restriction'
1158778197 M * Borg- they usualy do bind(INADDR_ANY,port) and do connect (127.0.0.1)
1158778207 M * Bertl i.e. you have a bunch of addresses on the host
1158778217 M * ray6 borg: its modifying the bind-systemcall to use just the correct IPs of that vserver
1158778226 M * Bertl and you assign (usually) a subset of those to each guest
1158778239 M * ray6 Borg: if they do connect(127.0.0.1) they're buggy, they should lookup "localhost" :)
1158778244 M * Borg- ray6: what about client side? configs are quiet sophosticated you know.. I can just go and change everywhere to 127.0.0.1
1158778248 M * Bertl the guest then will only be able to bind to those ips ...
1158778260 M * Borg- ray6: I know.. lazy bitches developers..
1158778264 M * ray6 Borg: so if the host is for network testing - that's a good test to find buggy programs :)
1158778266 M * Borg- but for now I cant do anything about that
1158778277 M * Bertl Borg-: no need to, first, as I said, 127.0.0.1 is remapped to the first ip
1158778287 M * Borg- Bertl: ahh.. thats good then.
1158778294 M * Bertl (which of course, could be 127.0.0.1 too :)
1158778301 M * ray6 oh, no finding of buggy programs then? :/
1158778304 M * ray6 :)
1158778316 M * Borg- thats all what I need :) but well.. everything will be clear tumorrow
1158778319 M * Bertl second, looking up localhost (which also should resolve to the first ip)
1158778330 M * Borg- btw.. im quite dissapointed that there is no Requirements for vserver on page..
1158778333 M * Bertl solves almost all other issues like security checks
1158778334 M * Borg- at least I couldnt find it easly.
1158778348 M * Bertl Borg-: what 'requirements' do you expect?
1158778352 M * Borg- not im stuck w/ compiling iproute2, iptables, nameif.. and who knows what else
1158778373 M * Bertl Borg-: well, basically you can do without all that
1158778382 M * Borg- ./configure bothers me..
1158778392 M * Borg- 2.6.17.13
1158778396 M * Bertl Borg-: you are now talking about 'specific' tools, right?
1158778416 M * Bertl i.e. util-vserver I presume
1158778428 M * Borg- yeah
1158778446 M * Borg- they are "official" ;) so I think I should use them as newbie
1158778447 M * Bertl well, those tools have certain requirements, taht is correct
1158778462 M * ray6 Borg: IIRC there's a source-RPM for the utils-vserver IIRC, from daniel or so, you could look into that spec file for build requirements :)
1158778488 M * daniel_hozac the spec file is in the tarball.
1158778488 M * Bertl yeah, but IIRC, the ./configure gives somewhat precise infos where to look
1158778492 M * daniel_hozac indeed.
1158778501 M * Borg- ray6: well.. for now im handle it well.. ./configure borks.. im looking web or sources on CD :) and compile by hand
1158778525 M * Borg- I just wonder how much of that is..
1158778537 M * Borg- this should be all listed in README..
1158778538 M * Bertl Borg-: what distro?
1158778547 M * Borg- Bertl: very old one ;) I dont want to tell.. hehe
1158778573 M * ray6 Borg: note your findings and add them to a page in the wiki? :) 
1158778588 M * Borg- ray6: hmm yeah.. I could do it
1158778593 M * Bertl yeah, or send it to the ML
1158778593 M * daniel_hozac Borg-: like REQUIREMENTS?
1158778615 M * Borg- daniel_hozac: yeah
1158778627 M * daniel_hozac there you go then.
1158778668 M * Bertl hehe, lol, yeah
1158778672 M * Bertl didn't see that one :)
1158778697 M * Bertl but I'm not sure it is up-to-date
1158778933 M * Nei Bertl, Borg-, 'jftpgw' sounds like a program that could benefit my desires
1158778989 M * Borg- Nei: I still dont get your problem :)
1158779028 M * Borg- you want to provide FTP access for your customers only? (so they can access web files.. etc)
1158779040 M * Nei Borg- single ip server, vservers on private virtual network, want to give out ftp accounts to users which works on server:21
1158779046 M * Borg- or you want to allow anyone to use your customer's ftp (anonymous one)
1158779066 M * Borg- because there is a big difference in those 2
1158779105 M * Nei the best thing would be if they can manage their ftp themselves
1158779143 M * Nei the first thing would be ftp access for them to manage their web files etc though, so your #1
1158779145 M * Borg- FTP active initiates connection from server to client.. so this is not a problem
1158779151 M * Borg- the bigger problem is w/ passive mode
1158779173 M * Nei the main problem for me was that I can only run one ftp on port 21
1158779210 M * Borg- yeah.. its quite complicated..
1158779225 M * Borg- first one is easy to do.. but second one.. I dont have clue for now
1158779301 M * ray6 nei: why do you want more than one ftp-server?
1158779319 M * Nei ray6 multiple vservers, multiple customers?
1158779328 M * ray6 nei: and you only have one IP?
1158779331 M * Nei Borg- how to do te first one?
1158779335 M * Nei ray6 exactly
1158779352 M * ray6 hm, normally ftp should be possible on other ports as well I guess?
1158779366 M * Nei yep
1158779381 M * Nei I was looking for a default port solution
1158779389 M * Bertl should be no problem to have a proxy for that
1158779394 M * ray6 for passive-ftp: pro-ftp can be assigned a port-range for the passive ports. Use 1000-1100 for your first and so on
1158779400 M * Bertl assign a private ip to each guest and simply forward
1158779423 M * Nei 'forward'?
1158779434 M * ray6 Bertl:  you'd need a reverse-ftp-proxy for that, don't know one
1158779434 M * Nei afaik there can only be one forwarding on a port ^^;
1158779438 M * Bertl well, the proxy has to do that
1158779447 M * ray6 Nei: can you diverge your users by there source IP?
1158779452 M * essobi_ So um... Any comments on installing an iso to vserver?
1158779459 M * Nei ray6 no way heh
1158779473 M * Nei essobi_ maybe you'd rather look into a real emulator
1158779479 M * ray6 Nei: thet you're lost. Give them non-default ports :)
1158779493 M * essobi_ Nei:  They said earlier it was possible.
1158779495 M * Nei vservers are more like  chroots on steroids
1158779501 M * Nei you cant run your own kernel in it
1158779510 Q * wdv Quit: Leaving
1158779512 M * essobi_ Nei, it all else fails, I'll just install it somewhere and tar up the whole install. :)
1158779519 M * ray6 esso: install it in vmware and then copy it into the vserver :) That works for all distros. For special distros therer are other ways
1158779536 M * Nei what iso is it?
1158779538 M * ray6 (or use a real server instead of vmware :) What distro is your ISO?
1158779548 M * Borg- Nei: pure-ftpd w/ Virtual passwd ? and you give accounts like: <login>@<domain>
1158779548 M * essobi_ PoundKey.
1158779555 M * essobi_ It's a minimalistic asterisk environment
1158779578 M * ray6 esso: and they don't offer an image? Or simply have an image+installer in the iso?
1158779581 M * Borg- this works quite nice.. also pure-ftpd have something like Virtual-Host stuff.. but I never really read what it does
1158779594 M * essobi_ It's an installer iso.
1158779598 M * Bertl Nei: maybe this one is of some use: ftp://ftp.suse.com/pub/projects/proxy-suite/src/
1158779599 M * ray6 then install it somewhere and build your own image with tar - should be no problem. 
1158779613 M * essobi_ That's what I was thinking..
1158779633 M * Nei Borg- so you mean running pure-ftpd on the host and chroot the user in his vserver dir?
1158779641 M * Borg- Nei: yep
1158779650 M * daniel_hozac essobi_: mount the ISO and see what kind of packages it's using?
1158779660 M * Nei that would be unable to cope with contexts and the vserver's user mappings though I think
1158779684 M * Bertl Nei: it would be possible but not really advised for security reasons
1158779694 M * ray6 ftp is sooooo 80s anyway. Use ssh/sftp to manage your webroots :)
1158779698 M * Nei yea and also security is a moot point
1158779707 M * Nei ray6 tell that to your customers ;)
1158779717 M * ray6 Nei: I do and it works :)
1158779718 M * Borg- hmm.. SFTP is an option :) but yeh.. customers usualy dont care ;)
1158779732 M * Nei great if yours have no problem 
1158779754 M * ray6 Nei: I know customers who you tell "you now also can use ftp+tls" and they ask "why don't you just offer sftp" :>
1158779760 M * Borg- ray6: done comparsion to 1Gbit SFTP vs FTP?
1158779790 M * Nei Bertl I wonder what that stuff is, I'll take a look
1158779809 M * ray6 Borg: I have a GBit upling for my servers, but I never had use to manage webroots over GBit - who has that much upstream bandwidth except for us? :)
1158779856 M * Bertl ssl allows for fast ciphers too .. you can easily get 50M/s and more
1158779881 M * Borg- ray6: right.. I was just wondering :)
1158779995 M * ray6 nei: or let your customers use openvpn and ftp through that to their private IP and tell them ftp isn't secure without that :)
1158780008 M * blizz is it possible to change an interface ip "on the fly" without a reastart?
1158780060 M * Borg- ray6: geez ;) they are just simple customers.. you know? ;]
1158780082 M * Nei ray6 hihi I fancy they'd be even more shocked than using winscp
1158780164 M * Borg- hmm quite nice network.
1158780168 M * Borg- (IRC network I mean)
1158780180 M * Nei how so?
1158780182 M * Borg- is there any operator/admin related channel? I wanna take a look before idling ;)
1158780208 M * Bertl check out #oftc
1158780221 M * Nei Bertl thanks again that description sounds indeed promising
1158780237 M * Bertl Nei: no idea how it actually is, just found it via google
1158780259 M * Nei :)
1158780265 Q * coocoon Quit: KVIrc 3.2.0 'Realia'
1158780278 M * Borg- Bertl: thx
1158780300 M * Nei maybe I'll find out ;)
1158780332 M * Bertl Nei: if it works for you, please drop us a note or even better write a wiki page :)
1158780666 M * Nei it'd be pretty cool to at least give back a little thing, but then again I'm lazy and an asshole en plus
1158780678 M * bronson Anyone use Ubuntu 6.06 LTS?
1158780690 M * Borg- Nei: lazy good.. asshole.. baaaaad
1158780696 M * Nei we'll see ;)
1158780706 M * bronson (under VServer?)
1158780715 M * bronson Or should I just go Debian like everyone else?  :)
1158780721 M * Borg- bronson: now everyone else
1158780727 M * Borg- im using diff ditro w/ vserver
1158780731 M * Borg- :P
1158780746 M * bronson diff?  haven't heard of that.  url?
1158780752 M * Borg- still I didnt finished the install ;) 
1158780759 M * Borg- bronson: diff == different distro :)
1158780784 M * phreak`` Bertl: working on a patch for 2.6.18 ? :)
1158780786 A * phreak`` runs
1158780803 M * Nei bronson ubuntu is based on the debian package system
1158780828 M * Nei personally I had not so much fun with it
1158780838 M * bronson True; it *should* work.
1158780850 M * bronson Nei, you've tried Warty under VServer?
1158780855 M * Bertl phreak``: almost done! :)
1158780866 M * daniel_hozac bronson: you'll need to install Ubuntu's debootstrap, or point util-vserver at it to install them.
1158780872 M * Nei it seems to be pretty popular, but I've read on quite a few blogs of people paddling back
1158780888 M * Nei daniel_hozac debian has ubuntu scripts
1158780893 M * daniel_hozac really?
1158780900 M * Nei at least mine does
1158780902 M * daniel_hozac must've missed that.
1158780909 M * Nei maybe not in sarge
1158781092 M * phreak`` Bertl: great :)
1158781589 M * Nei but debootstrap has almost no dependencies so it wouldnt be difficult in either case
1158781816 Q * lilalinux Remote host closed the connection
1158782086 M * Nei Bertl as I mentioned I also fount jftpgw which promises about the same thing as the suse proxy suite, so I'm currently trying to judge from their descriptions & docs which one I want to try (first)
1158782104 Q * Blissex Remote host closed the connection
1158782238 N * Piet__ Piet
1158782715 J * Piet_ hiddenserv@tor.noreply.org
1158782829 J * marcfiu ~mef@targe.CS.Princeton.EDU
1158782833 M * marcfiu Hi bertl.
1158782836 M * Bertl welcome marcfiu!
1158782854 M * marcfiu Your latest scheduler can do both fair sharing AND reservations, right.
1158782874 M * marcfiu Whereas Andy's scheduler could only do one or the other.
1158782880 M * Bertl yes, it should be a superset of what planetlab had
1158782903 M * marcfiu What did you do so you could support both reservations and fair sharing?
1158782917 M * Bertl at least Andy could not name any case which would not be mapable 
1158782934 M * Bertl basically we generalized the idea of the time-warp
1158783031 M * marcfiu The rate that tokens accumulate in a VM's bucket depends on whether
1158783031 M * marcfiu the VM has a {\it reservation} or a {\it share}.  A VM with a
1158783031 M * marcfiu reservation accumulates tokens at its reserved rate: for example, a VM
1158783031 M * marcfiu with a 10\% reservation gets 100 tokens per second, since a token
1158783031 M * marcfiu entitles it to run a process for one millisecond.  On PlanetLab, for
1158783032 M * marcfiu example, the default share for a VM is actually a small reservation,
1158783032 M * marcfiu providing the VM with 32 tokens every second, or 3\% of the total
1158783034 M * marcfiu capacity.
1158783034 M * marcfiu The main difference between reservations and shares occurs when there
1158783036 M * marcfiu are runnable processes but no VM has enough tokens to run: in this
1158783036 M * marcfiu case, VMs with shares are given priority over VMs with reservations.
1158783038 M * marcfiu First, if there is a runnable VM with shares, tokens are given out
1158783038 M * marcfiu fairly to all VMs with shares (i.e., in proportion to the number of
1158783040 M * marcfiu shares each VM has) until one can run.  If there are no runnable VMs
1158783040 M * marcfiu with shares, then tokens are given out fairly to VMs with
1158783042 M * marcfiu reservations.  The end result is that the CPU capacity is effectively
1158783042 M * marcfiu partitioned between the two classes of VMs: VMs with reservations get
1158783044 M * marcfiu what they've reserved, and VMs with shares split the unreserved
1158783044 M * marcfiu capacity of the machine proportionally.
1158783046 M * marcfiu This is the text I have for how Andy's scheduler works.
1158783074 Q * Piet Ping timeout: 480 seconds
1158783084 M * marcfiu What have you done to generalize time-warp?
1158783093 M * Bertl well, the current scheduler works like this:
1158783110 M * Bertl - tokens are added on a rate/interval basis
1158783131 M * Bertl - when the idle task would be scheduled, we switch into so called idle-time
1158783150 M * Bertl - then the scheduler advances idle time until a context can be scheduled again
1158783166 M * Bertl - this is done using a second rate'/interval' set for each guest
1158783207 M * Bertl - and before the idle task is scheduled, a context process is scheduled instead (unless no context has idle time enabled)
1158783352 M * Bertl that's it :)
1158783473 M * marcfiu Need to bend my head around rate/interval.
1158783504 M * Bertl ah, that's quite simple too:
1158783504 M * marcfiu Could you give me an example of how a context can have BOTH a fair share and a reservation.
1158783525 M * Bertl every interval jiffies/ticks, the rate amount of tokens is added to the bucket
1158783545 M * Bertl 100/25 means every 100 ticks, you get 25 tokens
1158783727 M * Bertl doener: I have a version for 2.6.18, but it throws strange warnings here: warning at kernel/fork.c:119/__put_task_struct()
1158783739 M * Bertl doener: what is your status?
1158783760 M * Bertl (seems to be related to the procfs changes)
1158784129 M * marcfiu bertl: ok
1158784220 M * marcfiu How does on get a L/M share (where M=|contexts|) and N% of the CPU?
1158784306 M * marcfiu E.g., can a context say it wants a fair share AND 10% of the CPU, such that it can get a fair share of the capacity available above it 10% reservation?
1158784317 M * marcfiu Obviously your answer is yes.
1158784342 M * marcfiu I just want to know how you expose that via the tokens.
1158784368 M * marcfiu Are there "fair share" tokens and "reservation" tokens, and a context is scheduled based on the sum of these two tokens?
1158784408 M * Bertl well, you give the 10% via the rate/interval (assumed that you do not overbook the system, this will be a guarantee)
1158784427 M * Bertl and you assign equal rate'/interval' values to each guest
1158784445 M * Bertl (which will give you a fair share for each of them)
1158784497 M * marcfiu ?
1158784516 M * marcfiu so there is just one rate/interval value per guest, right.
1158784517 M * Bertl as long as something is running, the idle time will not be touched
1158784529 M * Bertl marcfiu: nope, there are two pairs 
1158784556 M * Bertl one (rate/interval) for 'normal' time, and another one (rate'/interval') for the idle-time
1158784573 M * marcfiu what does '/interval' mean?
1158784587 M * marcfiu Is that your short hand for the idle-time ?
1158784607 M * marcfiu I think I am just trying to get my head around your terminology.
1158784608 M * Bertl well, the prime just denotes a different set
1158784612 M * marcfiu ok
1158784614 M * marcfiu understood
1158784631 M * Bertl make that rate[0]/interval[0] and rate[1]/interval[1] :)
1158784635 M * marcfiu yes
1158784638 M * marcfiu thank you.
1158784651 M * marcfiu the '/interval' confused me.
1158784670 M * Bertl well, it's r/i or rate per interval
1158784697 M * marcfiu I think Andy's implementation was to leverage just one rate/interval definition + a flag to indicate whether it was a guarantee or fair share.
1158784720 M * marcfiu whereas you have one rate/interval pair for both guarantee and fair share.
1158784725 M * marcfiu right?
1158784730 M * marcfiu (please say yes)
1158784741 M * marcfiu :)
1158784768 M * Bertl yeah, basically :)
1158784782 M * Bertl we have a flag to en/disable the idle-time per guest too
1158784787 M * marcfiu so the "normal time" rate/interval pair == guarantee, and
1158784787 M * marcfiu           "idle-time" rate/interval pair == fair share.
1158784805 M * Bertl and it is ot strictly a guarantee, as you can as well overbook the system
1158784818 M * marcfiu yes, I understand fully why you'd want to disable the idle-time per guest.
1158784822 M * Bertl but! it becomes a guarantee if you do not do that :)
1158784835 J * DreamerC_ ~dreamerc@61-217-227-68.dynamic.hinet.net
1158784836 M * marcfiu I.e., for hosting scenarios where a guest owner purchased 10% of the CPU and shouldn't get anymore.
1158784886 M * derjohn Bertl, ddi you have a look at the 2.6.18 changes?
1158784912 M * Bertl derjohn: we are on it
1158784964 M * derjohn Bertl, ah, didnt want to disturb ;) But .. as I already did: Are there heavy changes?
1158785048 M * Bertl yep, definitely
1158785207 M * marcfiu Bertl: suppose a context only has a rate/interval for idle-time.  When does it get to run?
1158785218 M * marcfiu Only when the idle thread would normally run?
1158785227 M * Bertl precisely
1158785234 Q * DreamerC Ping timeout: 480 seconds
1158785290 M * marcfiu So if the system is overbooked with contexts that have rate/interval, then those with just rate'/interval' would starve, right?
1158785305 M * Bertl exactly
1158785305 M * marcfiu That makes sense...
1158785306 M * derjohn hm, the recent kernel development model scares me more and more.
1158785353 Q * bonbons Quit: Leaving
1158785383 M * marcfiu However, if the aggregate of contexts with rate/interval is only, say, 50% of total system capacity, then the contexts with rate'/interval' would get the other 50% fairly?
1158785414 M * Bertl well, or according to the r'/i' values assigned
1158785418 M * marcfiu right
1158785425 M * marcfiu that's what I meant.
1158785430 M * Bertl i.e. you can divide them like 1:2:3 too
1158785435 M * marcfiu right
1158785438 M * marcfiu good
1158785460 M * marcfiu very nice
1158785470 M * marcfiu I think I am done pestering you. :)
1158785470 M * Bertl yes, indeed :)
1158785482 J * Piet hiddenserv@tor.noreply.org
1158785657 M * doener Bertl: got busy with other stuff here at home, not much progress yet
1158785696 M * Bertl np, I'll upload the first try shortly, but it crashes on ps inside a context
1158785747 Q * Borg- Quit: leaving
1158785894 Q * Piet_ Ping timeout: 480 seconds
1158786141 Q * Piet Remote host closed the connection
1158786194 M * Bertl doener: http://vserver.13thfloor.at/Experimental/patch-2.6.18-vs2.0.2.1-try1.diff
1158786217 M * Bertl doener: maybe you have an idea what happens when you do 'chcontext --xid 100 -- ps' :)
1158786303 J * harti ~Hartmut@85-124-97-58.dynamic.xdsl-line.inode.at
1158786311 M * Bertl welcome harti!
1158786317 M * harti hey
1158786585 M * Bertl hmm, hmm, I got a feeling I know what happens :)
1158786679 J * Piet hiddenserv@tor.noreply.org
1158786680 Q * comfrey Quit: Lost terminal
1158787205 Q * mire Quit: Leaving
1158787455 N * kaner_ kaner
1158787650 A * Belu is away (iŽll be back later...)
1158787651 N * Belu Belu_zZz
1158787816 J * mire ~mire@102-167-222-85.COOL.ADSL.VLine.verat.net
1158787858 Q * mire 
1158787949 Q * dna_ Read error: Connection reset by peer
1158788576 J * Aiken ~james@tooax6-210.dialup.optusnet.com.au
1158788943 M * Nei Bertl I found this with google http://swapoff.org/wiki/LinuxVServer
1158789030 J * dna_ ~naucki@223-250-dsl.kielnet.net
1158789061 J * mire_ ~mire@102-167-222-85.COOL.ADSL.VLine.verat.net
1158789470 Q * dna_ Quit: Verlassend
1158790086 M * Bertl Nei: nice ...
1158790166 M * doener Bertl: so, do you know now? :)
1158790179 M * Bertl yeah, this specific issue is fixed :)
1158790634 Q * harti Quit: Client exiting
1158791707 N * DreamerC_ DreamerC
1158792428 M * matti :)
1158792443 M * matti Almost in UK, uff.
1158792678 M * gdm see ya soon ;-)
1158792731 M * matti Oh, hello gdm :)
1158792753 M * gdm hehe... actually, am about to go to sleep
1158792770 M * gdm but just thought i'd say hi
1158792775 M * gdm nite all!
1158792820 M * matti gdm: I can meet you in London at 27/09 (wednesday).
1158792826 M * matti gdm: Well, sleep well mate :)
1158792834 M * matti ;p
1158792866 M * gdm matti: ok. send me an email or something. .. that's next week, right
1158792873 M * gdm we'll sort something out
1158792897 M * matti Yep. Next week.
1158792908 M * matti :)
1158793613 P * marcfiu 
1158793712 P * stefani I'm Parting (the water)
1158794053 M * Bertl running final test now ... looks good so far
1158794109 Q * Piet Quit: :tiuQ
1158794888 M * doener *sigh* arrived at fs/proc/base.c again ;)
1158795036 M * Bertl it's getting quite simple there, actually
1158795071 M * doener yeah, as I said, I just don't feel confident about it
1158795099 M * doener s/it/my knowledge about it/
1158795380 Q * id23 Ping timeout: 480 seconds
1158795703 M * Bertl http://vserver.13thfloor.at/Experimental/patch-2.6.18-vs2.0.2.1-t4.diff
1158795718 M * Bertl still labeled as testing, but it passed all my tests
1158795742 M * mnemoc t4?
1158795752 M * mnemoc ah, ok :)
1158795816 J * _node node@c-69-143-148-254.hsd1.md.comcast.net
1158796135 M * matti OMG
1158796144 M * matti I have missed the 2.6.18 ;D
1158796149 M * matti ;p
1158796420 J * FireEgl ~FireEgl@Sebastian.Atlantica.US
1158796448 M * matti Whooo... changelog for .18 looks very impressive ;]
1158796550 J * sladen_ paul@starsky.19inch.net
1158796550 Q * sladen Read error: Connection reset by peer
1158796696 Q * FireEgl Read error: Connection reset by peer
1158796765 M * Bertl matti: yep, it had quite some changes