1255738347 Q * geb Ping timeout: 480 seconds
1255738358 J * geb ~geb@earth.gebura.eu.org
1255738903 Q * bonbons Quit: Leaving
1255740811 J * mathx ~math@sizone.org
1255740830 M * mathx ohai. um. not finding much docs on vhashify. now that ive vhash'd a guest, how do i.. use it?
1255740856 M * mathx my goal is to take this image of a vserver and upgrade it to lenny from etch, but leave the original as is for benchmarking to compare to lenny
1255741615 Q * ViRUS Quit: If there is Artificial Intelligence, then there's bound to be some artificial stupidity. (Thomas Edison)
1255741615 M * Bertl http://linux-vserver.org/Vhashify
1255741637 M * Bertl http://linux-vserver.org/Frequently_Asked_Questions#How_do_I_manage_a_multi-guest_setup_with_vhashify.3F
1255741649 M * mathx right, i read the FAQ
1255741653 M * mathx followed the instrux
1255741669 M * mathx first url i didnt find, checking thx
1255741670 M * Bertl good, so now your guests are unified ...
1255741691 M * Bertl they will consume less disk space, less memory and run faster overall
1255741762 M * mathx well, .. ish. :) not really.
1255741766 M * mathx most of my sites are incredibly different.
1255741775 M * mathx some ubuntu, some lenny, some squeeze
1255741778 M * mathx the odd FC
1255741791 M * mathx but in this single case, it's the DATA i want hashed, not the libs and apps
1255741805 M * mathx because im about to make 2-3 different clones of this server and benchmark each with the changes i make
1255741814 M * Bertl well, wherever they share data, vhashify will 'combine' them
1255741820 M * mathx but i need the reference server in place. it has 750k files of user/app data in it, i dont wnat to duplicate that
1255741823 M * mathx right.
1255741833 M * mathx sharing data is uncommong. sharing APPS is more common.
1255741837 M * mathx in this case, the reverse.
1255741847 M * mathx sides, disk and ram are CHEAP
1255741870 M * mathx what's expensive is disk slots and statting 20M files for a daily backup
1255741881 M * mathx really need a real FS for linux :(
1255741887 M * mathx one with file alteration logs
1255741896 M * mathx why are we stuck in 2005 with linux filesystems?
1255741904 M * mathx (answer: licensing. not that ZFS has FA logs...)
1255741928 M * Bertl hmm, maybe want to try btrfs then?
1255741928 A * mathx has 49 vservers on one box
1255741933 M * mathx well, its BETA
1255741938 M * mathx more beta than R4
1255741949 M * mathx i have some R4 in places for customers that really need it and are in dev mode (not production)
1255741956 M * mathx not that R4 has FA logs either!
1255741978 M * mathx im _THIS CLOSE_ to caving in on fuse and trying out ZFS, but then again, Debian FreeBSD exists mainstream now
1255742000 M * Bertl I consider btrfs of much better quality already than reiserfs will ever get
1255742005 M * mathx (also i have to look at inotify + rsync = lsyncd, but i dont think it can track 20M files)
1255742013 M * mathx really? hmm.
1255742024 M * Bertl anyway, whatever you find suitable for your purpose ... go with it
1255742030 M * mathx the one place im using it is for a customer with an egregious model, they're creating 25M files avg size 130 bytes.
1255742041 M * mathx R4 is optimized for small files like that
1255742045 M * Bertl I'm off to bed now ... have fun!
1255742047 M * mathx but i should try brtfs, i compiled it
1255742051 M * mathx into that kernel
1255742052 M * mathx ok nute
1255742054 M * mathx *nite
1255742059 N * Bertl Bertl_zZ
1255742065 M * mathx zZ = zfs!
1255742083 A * mathx finishes his 3rd beer and stumbles off to meet the wife at the bus station
1255743023 M * sid3windr (@mark please do not crop mail
1255743024 M * sid3windr threads in the future
1255743028 M * sid3windr = ehh, what?!
1255743092 M * sid3windr I cannot imagine one prefers to store useless gigabytes of data of text that is already in other mail and not necessary for the reply at hand
1255743095 M * sid3windr :|
1255746951 Q * geb Quit: /
1255748436 J * saulus_ ~saulus@c193008.adsl.hansenet.de
1255748844 Q * SauLus Ping timeout: 480 seconds
1255748852 N * saulus_ SauLus
1255749903 J * saulus_ ~saulus@c207075.adsl.hansenet.de
1255750029 Q * SauLus Ping timeout: 480 seconds
1255750039 N * saulus_ SauLus
1255751657 J * Shinsaku ~Shinsaku@chello089076140236.chello.pl
1255754062 Q * hparker Quit: Read error: 104 (Peer reset by connection)
1255755423 Q * FireEgl Ping timeout: 480 seconds
1255755602 Q * pmenier Ping timeout: 480 seconds
1255755738 J * pmenier ~pmenier@ACaen-152-1-5-203.w83-115.abo.wanadoo.fr
1255757673 J * FireEgl Proteus@2001:470:e056:1:4::9
1255760863 Q * AndrewLee Ping timeout: 480 seconds
1255761236 J * AndrewLee ~andrew@u7.hlc.edu.tw
1255762178 J * hparker ~hparker@2001:470:1f0f:32c:290:96ff:fe50:40fa
1255765436 J * derjohn_mob ~aj@e180194156.adsl.alicedsl.de
1255765632 Q * ghislainocfs2 Ping timeout: 480 seconds
1255767735 J * hparker|laptop ~hparker@2001:470:1f0f:32c:290:96ff:fe50:40fa
1255767739 Q * hparker|laptop Remote host closed the connection
1255767819 J * ghislainocfs2 ~Ghislain@adsl2.aqueos.com
1255768854 M * ghislainocfs2 do you know if i make the tools search for /var/lib/vservers for the root directory where the guest are
1255768874 M * ghislainocfs2 and that /var/lib/vservers is a link to /vservers will it be an issue ?
1255769933 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d
1255770006 J * Acksyis ~Acksyis@ANantes-258-1-63-159.w90-49.abo.wanadoo.fr
1255770019 M * Acksyis Sallut les noob
1255770592 Q * Acksyis Ping timeout: 480 seconds
1255772057 N * Bertl_zZ Bertl
1255772061 M * Bertl morning folks!
1255772094 M * Bertl ghislainocfs2: IIRC, it is a little problematic, why would you want to do that?
1255772103 J * lucrus ~papo@host-84-223-100-32.cust-adsl.tiscali.it
1255772137 M * Bertl welcome lucrus!
1255773706 M * ghislainocfs2 bertl: this is for the package of the tools, if want it to be the more debian possible but i wanted also that it gracefully handle the /vservers case
1255773721 M * ghislainocfs2 bertl: as for debian i should go for /var/lib/vservers instead
1255773802 M * Alteisen .oO( Debian and it's crude ideas where to put stuff ;-)
1255773901 M * ghislainocfs2 bertl: i can try a hard link too, don't know of a good way to handle both case. I want to avoid creating 2 packages :(
1255773937 M * ghislainocfs2 bertl: right now i configure /var/lib/vservers and if i find /vservers i  ln -s /vservers /var/lib/vservers
1255773957 M * ghislainocfs2 bertl: but it seems you do not advise this way of doing things :)
1255774082 M * Bertl ghislainocfs2: a hard link for a directory would be interesting .. please do that :)
1255774181 M * Bertl I'd suggest to go with /vservers by default (thus providing 'normal' behaviour) and let the folks insisting on /var/lib/vservers put a link there (or mount the /vservers filesystem there too)
1255774271 M * Alteisen so /vservers is the usual directory for the vserver files as designed by the linux-vserver-developers, and debian is just defining other targets?
1255774281 M * ghislainocfs2 bertl lol okay...
1255774331 M * ghislainocfs2 Alteisen: debian rules forbid to create directory in / for packages (not a bad idea if not you will end with countless dir there)
1255774352 M * Alteisen aha
1255774356 M * ghislainocfs2 bertl: you will be surprise if i hardlinked a directory he !do NOT challenge me !
1255774358 M * ghislainocfs2 ;)
1255774382 A * Alteisen has a separate disk for the vservers, mountet as /vservers
1255774427 A * ghislainocfs2 use a lvm partition but mainly this is the same
1255774441 M * Alteisen ghislainocfs2: alternatvely, you might use /srv/vservers
1255774450 M * Alteisen or /opt/vservers
1255774467 M * ghislainocfs2 whatever the decision, the issue is to support the vservers way AND the debian one in one package
1255774505 M * Alteisen i dont like the vserver files in /var/lib/vservers - that sounds not logical to me
1255774549 M * Alteisen ghislainocfs2: are you building an alternative debian .deb for vserver-utils?
1255774572 M * ghislainocfs2 this is where the debian package put it, as well the mysql package put its database etc
1255774587 M * Bertl ghislainocfs2: you can't have both in one package
1255774608 M * ghislainocfs2 utils verser use /usr/local, debian use /usr. vservers use /usr/local/etc/init.d debian /etc/init.d etc...
1255774620 M * Bertl depending on how you configure util-vserver, new guests will be created in /vservers or /var/lib/vservers
1255774620 M * ghislainocfs2 bertl: how can i do that ?
1255774642 M * ghislainocfs2 bertl: i can tell the tool to look on both ?
1255774666 M * ghislainocfs2 bertl: or it is just that the tools will follow the links whataver they can be
1255774681 M * ghislainocfs2 bertl: i mean in /etc/vservers 
1255774683 M * Bertl there is no looking involved, existing guests will be linked to /vservers or /var/lib/vservers depending on their config
1255774696 M * ghislainocfs2 bertl: isse
1255774702 M * ghislainocfs2 bertl: i see 
1255774742 M * Bertl so, the config decision only affects new and _default_ configed guests, and you cannot have a default point to two directories anyways
1255774792 M * Alteisen i am using plain debian lenny packages, defaulting vdir to /usr/lib/vservers, but during installing a vserver guest, i can tell where it's root dir is, and i am using /vservers for that...
1255774803 M * Bertl ghislainocfs2: what you can do is write a post-install script, which goes through the configurations, guesses the guest locations and updates the links if the guest is in a different place
1255774861 M * Alteisen bvserver $VSERVERNAME build -m debootstrap --hostname $VSERVERNAME.$DOMAIN --interface $INTERFACE:$NETWORKIPADDRESS.$VSERVERIPADDRESS/$SUBNETMASK --context $VSERVERIPADDRESS --rootdir $VSERVERSROOTDIR -- -d $DEBIANRELEASE -m $DEBIANMIRROR -- --exclude=$EXCLUDEDPACKAGES --include=$INCLUDEDPACKAGES --arch=$ARCH
1255774892 M * Alteisen VSERVERSROOTDIR=/vservers
1255775078 M * ghislainocfs2 humm ok the config is just for creating the link in /etc/vserver it will not be used for anything else then ( i mean the /usr/local/etc/vservers/.defaults/vdirbase)
1255775141 M * ghislainocfs2 bertl: the --with-vrootdir=/var/lib/vservers just configure /usr/local/etc/vservers/.defaults/vdirbase ? if so yes i can easely  recreate it by looking at the directory
1255775183 M * Bertl yes, this and some other defaults) but you can check there
1255775466 M * ghislainocfs2 ok, i just have to see how in bash i can know if the symlink go to /vservers or /var/lib/vservers . I guess by the inode number*
1255775527 M * Bertl hmm? what about ls -la or stat?
1255775564 M * ghislainocfs2 i was wondering about silly links like ../../../../vservers instead of a plain stupid /vservers
1255775586 M * ghislainocfs2 stupid the otherway arround but you corrected
1255775646 M * ghislainocfs2 ok i will do it simple, complicated setup will not use my package lol
1255775798 M * ghislainocfs2 hum cd /usr/local/etc/vservers/.defaults/vdirbase; pwd  should do it :)
1255775802 A * Alteisen loves simple things
1255775831 A * ghislainocfs2 have KISS in mind as much as possible
1255775863 M * Bertl yeah, KISS isn't that bad :)
1255775903 M * Alteisen ghislainocfs2: how are you going to keep up with patches? debian is very good in patching and providing security updates...
1255775908 M * ghislainocfs2 bertl: this is just the dreamhost guy that provide kernel package and will host my package prefer a debienized version but all my setup is  vserver like created leading to some non KISS way of finding clues to make it work
1255775935 M * ghislainocfs2 i will try as i will use the packages for my internal setup
1255777256 M * ghislainocfs2 bertl: here is what i do: preinst: go to vdirbase, if it exist and copy it to .orig.  postinst: copy back .orig to vdirbase.  This should keep the used settings live without breaking vserver
1255777291 M * Bertl make sure to test that :)
1255777316 A * Alteisen can live with the debian binaries, but avoids the vserver-debiantools package
1255777382 M * Alteisen so i just have a debian maintained binary package, and my own simple scripts using the util-vserver commands
1255777401 A * ghislainocfs2 wonder what this "test" word is
1255777462 M * Alteisen ghislainocfs2: you are also known as the "unknown stuntman"? ;-)
1255777663 M * Bertl okay, off for now, have to do some shopping ...
1255777671 N * Bertl Bertl_oO
1255780710 J * doener ~doener@i59F5B216.versanet.de
1255780813 Q * doener_ Ping timeout: 480 seconds
1255781340 Q * derjohn_mob Remote host closed the connection
1255781495 N * DoberMann[ZZZzzz] DoberMann
1255782216 J * derjohn_mob ~aj@e180194156.adsl.alicedsl.de
1255783403 Q * Shinsaku 
1255785839 J * disposable disposable@blackhole.sk
1255785953 M * disposable does anybody have a packaged amd64 vserver-enabled 2.6.30 kernel for debian squeeze that i can download so that i don't have to reinvent the wheel?
1255786140 N * Wonka_ Wonka
1255786190 Q * kiorky Quit: leaving
1255786251 Q * hparker Quit: Read error: 104 (Peer reset by connection)
1255786251 J * kiorky ~kiorky@cryptelium.net
1255787822 Q * Loki|muh Remote host closed the connection
1255787848 J * Loki|muh loki@satanix.de
1255787973 Q * Loki|muh Remote host closed the connection
1255788043 Q * disposable Quit: leaving
1255788146 J * Loki|muh loki@satanix.de
1255788437 Q * nenolod Read error: Operation timed out
1255789279 J * nenolod ~nenolod@petrie.dereferenced.org
1255789910 J * crushedhat_ ~tim@sheba-ipv6.crushedhat.com
1255789910 Q * crushedhat Read error: Connection reset by peer
1255789911 N * crushedhat_ crushedhat
1255789941 N * Bertl_oO Bertl
1255789948 M * Bertl back now ..
1255792210 J * grharry ~root@ppp-94-65-226-149.home.otenet.gr
1255792380 M * Bertl time for a nap .. bbl
1255792383 M * grharry Hi, I am trying to run asterisk on a vhost, in order to access the v0 /dev/mISDN   will a plain mknod from v0 /path/vserver/dev/mISDN c xx xx will suffice ??? 
1255792403 M * Bertl v0 = host, yes
1255792445 M * Bertl depending on the actual actions the guest will do, you might need one or the other capability too
1255792461 M * grharry Bertl: tnx ... do I have to reconstruct the hole /dev/tree  ??? 
1255792487 M * Bertl I wouldn't suggest to do so ... as you would greatly reduce the security 
1255792506 M * Bertl guest root is able to access all devices present in /dev
1255792532 M * Bertl (well, actually all devices present on a device enabled partition)
1255792577 M * grharry Thanks I came across this  http://www.telephreak.org/papers/vpa/  and it states delete first and then re-create  which I didn't understand quite well ... 
1255792580 M * Bertl but you can simply copy (cp -va) the device nodes from the host, when you need them
1255792627 M * grharry G thanks  a million Bertl !!! :-) 
1255792643 M * Bertl you're welcome!
1255792665 M * grharry Kinda newbie here 
1255792686 M * Bertl not a problem, see topic :)
1255792930 P * grharry 
1255792956 M * Bertl nap now ... :)
1255792962 N * Bertl Bertl_zZ
1255793079 J * mrfree ~mrfree@host-62-10-66-75.cust-adsl.tiscali.it
1255794141 J * geb ~geb@earth.gebura.eu.org
1255794377 J * vServer_User ~vServer_U@host90-152-0-28.ipv4.regusnet.com
1255796677 Q * nou Ping timeout: 480 seconds
1255797263 Q * manana Ping timeout: 480 seconds
1255797853 J * manana mayday_206@84.17.25.144
1255799219 Q * mrfree Ping timeout: 480 seconds
1255802652 N * Bertl_zZ Bertl
1255802655 M * Bertl back again ...
1255803682 M * daniel_hozac Bertl: my friend with the crashing kvm, traced down the problem
1255803727 M * Bertl ah, good, what is/was it?
1255803730 M * daniel_hozac to the task_lock in task_get_vx_info called from kernel/posix-timers.c
1255803754 M * Bertl hmm, I remember having fixed something in this area?
1255803772 M * daniel_hozac yeah, we added that code to make sure the timers were executed with the right context.
1255803793 M * daniel_hozac but we need to not do task_lock/task_unlock, or it might deadlock.
1255803834 M * Bertl sec
1255803900 M * Bertl okay, we are doing the task_lock/unlock there, that's bad?
1255803905 M * daniel_hozac yeah.
1255803932 M * Bertl hmm, how so?
1255803965 M * daniel_hozac the task can already be locked when the timer is called.
1255803989 M * Bertl and the complete signalling path does not lock the task?
1255804070 M * daniel_hozac apparently not.
1255804076 M * Bertl I mean, we can safely grab the vxi from the task without locks in this case .... I think
1255804094 M * Bertl it won't really go away
1255804130 M * daniel_hozac yeah
1255804165 M * Bertl so we change that to get_vx_info()
1255804758 M * Bertl could you/your friend test if that fixes the issue
1255804761 M * Bertl ?
1255804821 M * daniel_hozac yeah.
1255811639 J * hgh ~hghghg@04ZAABKQH.tor-irc.dnsbl.oftc.net
1255811666 M * hgh Women's rights activists got my free-software programs deleted from sourceforge (and persued the deletion of my accounts everywhere else) along with my ~40 nexuiz videogame maps and other GPLd media (https://mikeeusa.pressword.com has the info and links to them gloating). They are also trying to have me imprisoned for my anti-women's rights beliefs.
1255811763 M * Bertl well, that sounds about right to me ... 
1255811807 M * Bertl although imprisoning might be a little extreme .. but you definitely should change your beliefs
1255811896 M * Bertl soo, anything Linux-VServer related?
1255812004 M * Marillion Bertl: is a spamer, not really person for asking
1255812097 M * trippeh But why would it be right, there is a fundamental thing called freedom of speech... Even going after someone and get their shit cut off/deleted, is stepping over the line.
1255812198 M * hgh Here is where they try to track me down and talk of having me arrested and how they have reported me to their local police and to the fbi:
1255812200 M * hgh http://geekfeminism.org/2009/10/08/psa-mikeeusas-hate-speech-and-harassment/
1255812202 M * hgh http://ubuntulinuxtipstricks.blogspot.com/2009/10/attention-folks.html
1255812208 M * hgh trippeh, they're trying to lock me up too
1255812226 M * hgh so no freedom of not just speech; no freedom at all
1255812281 F * ChanServ +o Bertl
1255812331 Q * hgh Killed (Ganneff (No reason))
1255812391 J * uva bno@118-168-238-45.dynamic.hinet.net
1255812497 J * Genghis ~Genghis@ph34r.my.d-n-s.org.uk
1255812522 N * Genghis Guest1461
1255812576 J * MooingLemur ~troy@shells195.pinchaser.com
1255812602 J * pinochle_ ~pinochle@adsl-76-193-171-186.dsl.chcgil.sbcglobal.net
1255812626 Q * Guest999 synthon.oftc.net oxygen.oftc.net
1255812626 Q * MooingLe1ur synthon.oftc.net oxygen.oftc.net
1255812626 Q * pinochle synthon.oftc.net oxygen.oftc.net
1255812626 Q * uva_ synthon.oftc.net oxygen.oftc.net
1255812626 Q * infowolfe_ synthon.oftc.net oxygen.oftc.net
1255812634 Q * balbir_ Ping timeout: 480 seconds
1255812800 J * infowolfe_ ~infowolfe@c-76-105-242-186.hsd1.or.comcast.net
1255813149 M * trippeh Hmmm.
1255813305 J * balbir_ ~balbir@122.172.5.0
1255814110 N * DoberMann DoberMann[PullA]
1255814316 M * trippeh Harassment and threats, maybe imprisonment isn't such a bad idea after all.
1255814318 M * trippeh ;-)
1255814594 M * Wonka were the nexuiz maps any good? and the software?
1255814597 M * Wonka *duck*
1255814696 M * Marillion trippeh: for spam too ;) well is offtopic, I'm calm now 
1255815092 J * blues ~blues@agh243.neoplus.adsl.tpnet.pl
1255815210 Q * blues_ Ping timeout: 480 seconds
1255816410 J * scientes ~scientes@174-21-161-91.tukw.qwest.net
1255817619 J * BenG ~bengreen@94-169-110-10.cable.ubr22.aztw.blueyonder.co.uk
1255817748 Q * lucrus Remote host closed the connection
1255818964 Q * ghislainocfs2 Quit: Leaving.
1255821032 Q * bonbons Quit: Leaving
1255823520 J * mrfree ~mrfree@host-62-10-66-75.cust-adsl.tiscali.it