1305508944 J * ViRUS_ ~mp@p54866B48.dip.t-dialin.net
1305509378 Q * ViRUS Ping timeout: 480 seconds
1305514106 Q * Romster Quit: Geeks shall inherit properties and methods of object earth.
1305514413 J * Romster ~romster@202.168.100.149.dynamic.rev.eftel.com
1305515021 J * bsingh ~balbir@122.172.22.61
1305517284 M * Bertl off to bed now ... have a good one everyone!
1305517290 N * Bertl Bertl_zZ
1305518524 J * derjohn_mob aj@tmo-074-95.customers.d1-online.com
1305525213 Q * ntrs_away synthon.oftc.net weber.oftc.net
1305525213 Q * nkukard synthon.oftc.net weber.oftc.net
1305525213 Q * tolkor synthon.oftc.net weber.oftc.net
1305525213 Q * eja synthon.oftc.net weber.oftc.net
1305525213 Q * FireEgl synthon.oftc.net weber.oftc.net
1305525492 J * Pogs241 ~user@cpe-65-186-205-111.insight.res.rr.com
1305525492 J * eja ~user@75.110.195.31
1305525492 J * FireEgl ~FireEgl@173-16-9-3.client.mchsi.com
1305525492 J * tolkor ~rj@tdream.lly.earlham.edu
1305525492 J * nkukard ~nkukard@41-133-112-179.dsl.mweb.co.za
1305525492 J * ntrs_away ~ntrs@vault08.rosehosting.com
1305525506 Q * bsingh Ping timeout: 480 seconds
1305525863 J * ncopa ~ncopa@3.203.202.84.customer.cdi.no
1305527485 P * kir1 Leaving.
1305527778 J * ghislain ~AQUEOS@adsl2.aqueos.com
1305528189 Q * FireEgl Read error: Connection reset by peer
1305528561 Q * derjohn_mob Read error: Connection reset by peer
1305529166 J * derjohn_mob aj@tmo-049-46.customers.d1-online.com
1305529246 Q * LcKuroro Quit: leaving
1305530065 Q * derjohn_mob Ping timeout: 480 seconds
1305530432 J * cuba33ci_ ~cuba33ci@111-240-167-186.dynamic.hinet.net
1305530784 Q * cuba33ci Ping timeout: 480 seconds
1305530790 N * cuba33ci_ cuba33ci
1305531168 J * FireEgl FireEgl@FireEgl.Com
1305531916 J * derjohn_mob ~aj@213.238.45.2
1305534079 J * BenG ~bengreen@cpc12-aztw24-2-0-cust146.aztw.cable.virginmedia.com
1305537741 J * bsingh ~balbir@122.172.22.61
1305537806 M * josef_ ensc: I've been in contact with micah@debian, regarding helping out to get the latest util-vserver into debian sid. Have you done anything against debian lately?
1305543108 J * ensc|w ~ensc@www.sigma-chemnitz.de
1305543715 N * Bertl_zZ Bertl
1305543719 M * Bertl morning folks!
1305543720 Q * _are_ Ping timeout: 480 seconds
1305543792 M * Bertl josef_: what do you mean by 'against debian'?
1305545359 J * _are_ ~quassel@vs01.lug-s.org
1305545483 M * josef_ Bertl, I'll go through differences in patches etc in a moment, so what I mean is that if any cooperation or likewise has been done, and if so, what that has amended to.
1305545638 M * daniel_hozac AFAIK the only differences are related to less than stellar policies in Debian.
1305545651 M * Bertl .deb packages can be directly built from the util-vserver source, and micah is around most of the time here ... besides strange debian policy, nothing is stopping a brand new release
1305545863 M * josef_ Well let's get to it then ;-)
1305546590 Q * BenG Quit: I Leave
1305546853 J * manana ~mayday090@84.17.25.149
1305548705 Q * manana Ping timeout: 480 seconds
1305551017 J * Piet ~Piet__@659AABP7J.tor-irc.dnsbl.oftc.net
1305551145 J * Roelke ~Roelke@95-36-36-169.dsl.alice.nl
1305551867 J * Piet_ ~Piet__@659AABP7T.tor-irc.dnsbl.oftc.net
1305552187 Q * Piet Ping timeout: 480 seconds
1305552625 N * Piet_ Piet
1305553986 J * TheSeer ~theseer@border.office.ballyhoo.de
1305554001 M * TheSeer heya :)
1305554370 M * TheSeer I'm having some serious problems creating a new vserver as it seems :-/
1305554408 M * TheSeer vserver foo build -m yum --hostname foo.local -- -d centos5
1305554450 M * TheSeer it keeps isntalling stuff just fine, but in the end i get some errors like "/var/tmp/rpm-tmp.54279: line 5: mkdir: command not found" 
1305554497 M * TheSeer it lists like packages it pretends to have installed and exits off without actually having creating anything i can see
1305554563 M * TheSeer anyone having any idea as to why that might happen?
1305554570 M * TheSeer The already existing vserver work just fine...
1305554737 M * daniel_hozac paste the entire output
1305554829 M * TheSeer found it...
1305554835 M * TheSeer well, partly at least
1305554850 M * TheSeer i removed the rpmforge repository from the default
1305554871 M * TheSeer and now it works
1305554891 M * TheSeer so might be a broken package in there that caused the build to die off early
1305554921 M * TheSeer so, nevermind...
1305555301 M * Bertl nap attack ... bbl
1305555316 N * Bertl Bertl_zZ
1305556898 J * hparker ~hparker@2001:470:1f0f:32c:beae:c5ff:fe01:b647
1305557544 J * harobed ~harobed@pda57-1-82-231-115-1.fbx.proxad.net
1305557704 Q * Pogs241 Quit: Pogs241
1305558188 J * dowdle ~dowdle@scott.coe.montana.edu
1305560639 Q * derjohn_mob Ping timeout: 480 seconds
1305562161 J * bonbons ~bonbons@2001:960:7ab:0:c81e:5b44:b241:4e55
1305562553 M * Roelke My hoster has given my multiple TUN devices, so i can use more then 1 client at a time, is it possible to use them all together at a port ? 
1305562838 Q * harobed Ping timeout: 480 seconds
1305564321 Q * TheSeer Quit: Client exiting
1305564931 J * petzsch ~markus@dslb-092-078-225-050.pools.arcor-ip.net
1305565645 Q * bsingh Ping timeout: 480 seconds
1305566538 J * bsingh ~balbir@122.172.22.61
1305570618 N * Bertl_zZ Bertl
1305570636 M * Bertl back now ...
1305570676 M * Bertl Roelke: sure, each IP can use the same port separately if done properly (i.e. not binding to *)
1305570705 M * Bertl Roelke: but you do not necessarily need more than one TUN devices to connect to more than one client
1305571266 J * manana ~mayday090@84.17.25.149
1305571657 M * Roelke how can i use multiple clients one 1 TUN device then ? 
1305571662 M * Roelke me hoster said it was not possible 
1305571905 M * Bertl check the irc logs from last week or so, we had that discussion before, basically it boils down to using different networks for server and each client and handling the connection via routing
1305572118 J * Roelke_ ~Roelke@95-36-36-169.dsl.alice.nl
1305572122 N * ensc Guest1154
1305572131 J * ensc ~irc-ensc@p5DF2F3CC.dip.t-dialin.net
1305572379 Q * sigue Read error: Connection reset by peer
1305572465 Q * Roelke Ping timeout: 480 seconds
1305572516 M * Roelke_ i've searched the logs, but couldn't find much info 
1305572543 Q * Guest1154 Ping timeout: 480 seconds
1305572735 J * sigue contempt@stole.ur.cc-number.info
1305572739 M * Bertl http://irc.13thfloor.at/LOG/2011-05/LOG_2011-05-09.txt
1305572752 M * Bertl follow the bread crumbs back to the origin
1305572843 M * Roelke_ you meant this one : http://linux-vserver.org/OpenVPN
1305572869 M * Roelke_ because all the other things from milenco was the same problem as i had 
1305572956 M * Bertl check http://irc.13thfloor.at/LOG/2011-05/LOG_2011-05-10.txt as well
1305572993 Q * sigue Read error: Connection reset by peer
1305573115 M * Bertl I'm not going to explain it over and over, especially as it isn't Linux-VServer related at all, if you need help, you can hire me (or somebody else) to help you with the setup ...
1305573138 M * Roelke_ uhu ok, no problem
1305573166 M * Bertl but separate tun devices work as well, of course
1305573187 M * Roelke_ that's working yes, but not on the same port
1305573203 M * Bertl then you are not binding to a single IP
1305573331 M * Bertl i.e. if you want different streams on the same port, you naturally need to have different IPs to bind to, if you want a single IP/port to handle many streams, you need a more sophisticated openvpn setup
1305573358 M * Roelke_ ok
1305573551 M * bonbons Bertl: how much work would it be to may sync be local to current namespace?
1305573618 M * Bertl huh?
1305573649 M * bonbons e.g. in order to prevent a guest that syncs (be it on restart/stop or at another opportunity) to cause dirty pages from unrelated mountôints (mountpoints not visible to the guest's namespace) from being flushed to disk
1305573691 M * bonbons e.g. if on host I have a process dirtying lots of files I don't want a guest restart/stop to flush files from host for a matter of performane
1305573694 M * Bertl ah, now I understand, not sure that sync doesn't do that already
1305573734 M * Bertl note that the 'host' usually is part of the guest namespace as well
1305573760 M * bonbons I had a look at kernel source but looks like normal sync just iterates over all superblocks (though there is a kernel side sync function that can take a superblock as input)
1305573966 M * bonbons not here, namespaces are at mostly cleaned from host's mountpoints
1305574036 M * Bertl what function did you check?
1305574114 M * bonbons sync_filesystems() in fs/sync.c
1305574205 J * sigue contempt@stole.ur.cc-number.info
1305574210 M * Bertl well, that one iterates about the superblocks, yes
1305574236 M * Bertl so it would be necessary to have a function which iterates over the mounts and gets the superblocks from there
1305574251 M * Bertl might result in duplicate syncs to those superblocks though
1305574348 M * bonbons or maybe do it the other way around, for each superblock check if it's mounted in current namespace (unless root/spectator namespace)?
1305574477 M * Bertl that can easily create a lot of kernel work/load
1305574533 M * Bertl i.e. you probably want to avoid that, but I think the duplicate sync should not hurt, as it won't cause any work on superblocks already in sync
1305574812 M * bonbons if sync is quick it will most probably still be clean when seen the second time, if things are getting dirtied quickly maybe not (though in that case sync is a pain anyhow)
1305574886 M * Bertl correct, so I'd take that approach, and you can do one simple optimization without adding more overhead: you can check the previous superblock to avoid calling sync on the same superblock twice in a row
1305575309 M * bonbons do you have a quick pointer as to where I should start looking in order to iterate over current's mountpoints? (proc for mounts/mountinfo or is there a better place?)
1305575329 M * Bertl the namespace
1305575375 M * Bertl it is referenced in the context
1305575472 M * Bertl struct mnt_namespace
1305575572 M * daniel_hozac see iterate_mounts in fs/namespace.c
1305575575 M * Bertl walk the vfsmounts and sync each super there
1305575583 M * daniel_hozac it has a callback
1305575586 M * daniel_hozac you can likely use it as is.
1305575602 M * bonbons thanks daniel, that's the area I was digging into
1305575604 M * Bertl without the twice in a row optimization, yes
1305575622 M * daniel_hozac it's got a void *, you can stash a list there.
1305575894 M * bonbons thanks for the hints (I guess I can do the patch against mainline so it benefits to everyone)
1305575912 M * Bertl yep, good idea
1305575964 Q * bsingh Ping timeout: 480 seconds
1305576390 J * derjohn_mob ~aj@d004192.adsl.hansenet.de
1305577265 J * hijacker_ ~hijacker@87-126-142-51.btc-net.bg
1305578482 J * Pogs241 ~user@cpe-65-186-205-111.insight.res.rr.com
1305578582 Q * manana Remote host closed the connection
1305578777 Q * bonbons Quit: Leaving
1305579539 Q * Roelke_ 
1305579645 J * derjohn_foo ~aj@d045021.adsl.hansenet.de
1305579664 Q * derjohn_mob Read error: Connection reset by peer
1305580640 Q * hijacker_ Quit: Leaving
1305580780 Q * petzsch Quit: Leaving.
1305582128 Q * Piet Remote host closed the connection
1305582172 J * Piet ~Piet__@659AABQEW.tor-irc.dnsbl.oftc.net
1305584694 Q * ghislain Quit: Leaving.
1305586420 Q * dowdle Remote host closed the connection
1305589207 J * ksn ~ksn@197.168.226.67