1113350436 Q * sebd Read error: Operation timed out
1113350453 J * sebd ~sebd@81.56.247.131
1113351301 M * Vudumen hmm :) i made consoles for the vservers :)
1113351304 M * Vudumen on tty7/8/9:)
1113351324 M * Bertl purpose?
1113351369 M * Vudumen i'm just playing with vserver :) i have this machine. it will be a real server soon but know i have time tonight :)
1113351395 M * Bertl ah, okay, well, then play on ;)
1113351410 M * Vudumen is there any drawbacks of this? security problem or other?
1113351429 M * Bertl no not really unless you use them from more than one vserver
1113351430 J * ciphernaut ~a@61.88.18.130
1113351450 M * Vudumen it's 3 different vserver :)
1113351621 M * Bertl evening ciphernaut!
1113351621 M * Vudumen Bertl: i don't see the mounts outside the vservers (/proc, /dev/pts). is it because of namespaces?
1113351636 M * Bertl yep should be so ...
1113351957 M * Vudumen and can i mount the root of the virtual fs in this namespace (with fstab or fstab.local)? now i use LVM volumes for the vservers.
1113352076 M * ciphernaut Hi all
1113352083 M * Vudumen hi ciphernaut 
1113352084 M * ciphernaut Hi Bertl
1113352255 M * albeiro Bertl: one thing, are you using vim for kernel developing ? if so, how have your configured it to work with hardlink trick ?
1113352382 M * Bertl yes, I do so .. sec I'll check my .vimrc ;)
1113352465 M * albeiro it's blazingly fast to diff / create new branches this way - make complete diff with another big project took... 5 seconds !
1113352474 M * Bertl set writebackup
1113352474 M * Bertl set backupcopy=no
1113352474 M * Bertl set patchmode=.orig
1113352485 M * Bertl guess that's the relevant part ...
1113352495 M * albeiro what is patchmode ?
1113352502 M * Bertl albeiro: what you don't say? ;) 
1113352553 M * albeiro if what touches my previous statement it would be rsbac ;]
1113353236 Q * alex234 Read error: Connection reset by peer
1113353323 M * albeiro http://vimdoc.sourceforge.net/htmldoc/options.html#'backupcopy' <- there is also breakhardlink option doing more or less the same.
1113353357 M * albeiro thx for above options, it will save me much disk trashing
1113353374 M * Bertl you're welcome!
1113353610 J * alex234 new@pD9E1FCF0.dip.t-dialin.net
1113353792 P * alex234 
1113354513 M * ciphernaut I have wriiten a script that brings my vserver into the state that I want it at.  Sarge on sarge setup.  I get funny locale problems.
1113354516 M * ciphernaut http://deadbeefbabe.org/paste/276
1113354604 M * ciphernaut has anyone else with a debian setup encountered something like this?
1113354998 M * Bertl hmm, I'd try to specify the following _before_ you do your script
1113355005 M * Vudumen ciphernaut: you mean when dpkg sais that locale is not found?
1113355012 M * Vudumen in my deploy script i unset LANG
1113355013 M * Bertl LC_ALL=POSIX
1113355017 M * Bertl LANG=C
1113355030 M * Vudumen i use this:
1113355030 M * Vudumen unset LANG
1113355030 M * Vudumen unset LANGUAGE
1113355044 M * Vudumen Bertl: what is the hideinfo flag for?
1113355073 M * Bertl which VXF does it correspond to?
1113355094 M * Bertl ah, already found it, sec
1113355161 M * ciphernaut Vudumen: pkg, yes
1113355176 M * ciphernaut unsetting works
1113355187 M * ciphernaut thanks all
1113355326 N * kevinp kevinp|gone
1113355532 M * Bertl Vudumen: it hides vxi and nxi information in the proc
1113355561 M * Bertl vxi: BCaps, CCAPs, CFlags, CIPid
1113355583 M * Bertl nxi: V4Root, V4RB
1113355795 M * Bertl Vudumen: vxc --xid 100 -- cat /proc/self/vinfo
1113355865 M * Vudumen thanks :)
1113355880 M * Vudumen hmm. what is vxc? :)
1113355890 M * Vudumen vcontext?
1113355898 M * Bertl a script to wrap the tools
1113355953 M * Bertl http://vserver.13thfloor.at/Experimental/TOOLS/vxc
1113355982 M * Vudumen thanks
1113355985 M * Vudumen again :)
1113355989 M * Bertl np
1113356000 M * Bertl vxc --xid 100 --flag ^5 -- cat /proc/self/vinfo
1113356010 M * Bertl (for example will not show that info)
1113356023 M * Vudumen agency:~# vxc --xid 201 --flag ^5 -- cat /proc/self/vinfo
1113356023 M * Vudumen vcontext: vc_create_context(): File exists
1113356023 M * Vudumen ret=254
1113356026 M * Vudumen is it okay?
1113356045 M * Bertl looks like you already have a context 201 ;)
1113356059 M * Vudumen yes i have :)
1113356068 M * Bertl use a number you don't have there ;)
1113356114 M * Vudumen it sais XID: 100
1113356145 M * Vudumen can i completely hide vserver infos in /proc in vservers?
1113356207 M * Bertl actually that was the idea of that flag, but it got lost on the way somehow ...
1113356232 M * Bertl if you are willing to do some testing, I can fix up this 'feature'
1113356255 M * Vudumen i'd like :)
1113356269 M * Vudumen until this machine goes to 100% production i can use it
1113356404 M * Vudumen and i found some virt_flags but i dodn't found any doc about them :(
1113356410 M * Vudumen virt_mem virt_uptime virt_cpu virt_load
1113356422 M * Vudumen now i know what is virt_uptime but the others?
1113356437 M * Bertl virt_cpu is not implemented yet
1113356447 M * Bertl virt_load gives you virtualzed load values
1113356469 M * Bertl and virt_mem uses the memory limits to generate some 'fake' memory values
1113356486 M * Bertl (probably broken as the /proc/meminfo format changed recently)
1113356577 M * Vudumen i see
1113356624 M * Bertl okay, what kernel/patch do you use right now?
1113356648 M * Vudumen 2.6.11.7-vs1.9.5.x
1113356653 M * Bertl excellent!
1113356657 M * Vudumen :)
1113356741 M * Vudumen someday i'll learn c at this level you know :)
1113356777 M * Bertl good, then you can help me ...
1113356784 M * Bertl (with the coding)
1113356793 M * Vudumen :)
1113356864 M * Bertl most folks underestimate the importance of testing and documentation
1113356905 M * Bertl usually it takes me a week to test a feature, five hours to reproduce an issue, 20 minutes to fix a bug ...
1113356938 M * Bertl so if I can concentrate on the bug fixing part ... it scales like 1:1000
1113356952 M * Vudumen i try it
1113356976 M * Vudumen about 2 years ago i tested vlc many times and fenrir fixed the bugs i found
1113357000 M * Vudumen we worked many times on the asf muxer
1113357052 Q * alexx Ping timeout: 480 seconds
1113357089 J * alexx ~alexx@82.225.136.176
1113357100 M * Bertl Vudumen: yeah, that is an excellent combination
1113357135 M * Bertl not to mention that some folks are really talented at finding bugs ...
1113357155 M * Bertl it's probably the way they think and try out things ...
1113357196 M * Vudumen yes. mostly i work on production systems and these systems can produce a lot of logs
1113357230 M * Bertl okay, first part is fixed .. now to the second ...
1113357264 M * Bertl (the information in /proc/self/status)
1113357370 M * Vudumen as i said at my workplace we are looking for a virtualization solution. currently we tested ensim (what was not a solution) and virtuozzo. virtuozzo seemed to be interesting but it's gui was too slapdashed and it didn't worth the money
1113357392 M * Bertl really?
1113357397 M * Vudumen yes
1113357415 M * Vudumen and this idiot thing that it should be used on redhat or suse...
1113357415 M * Bertl always thought virtuozzo was so good because of the gui?
1113357433 M * Vudumen not really but we needed it for other colleagues
1113357460 M * Vudumen we had other problems too. some things didn't worked as expected and everything was limited :/
1113357554 M * Vudumen but the most important is that it costs thousands of dollars and didn't have as manz advanteges as we think that it should if we pay for it when other soulrtions are available freely
1113357561 M * Vudumen (vserver, uml, xen, etc.)
1113357593 M * Vudumen the next what we will test will be VMware GSX server. it seems to be very interesting but i think GSX is much more what we need
1113357614 M * Vudumen and i decided that on my private server i'll use vserver :)
1113357639 M * Bertl excellent choice ;) (but I'm probably biased ;)
1113357653 M * Vudumen :)))
1113357656 M * Vudumen i think so :)
1113357795 M * Bertl yeah, but just a little bit ...
1113357925 M * Bertl okay, seems like second part is done too .. now the directory listing
1113357925 M * Vudumen little bit what? :)
1113357955 M * Bertl biased of course ;)
1113357961 M * ComplexHo Vudumen, I have some negative experiences with Virtuozzo too...
1113357972 M * Bertl let's hear!
1113358009 M * Bertl and please do not keep behind the bush with things to improve on linux-vserver ...
1113358046 M * ComplexHo Have a very close friend working at a large multinational hosting co who rolled out VIrtuozzo last year.  It's a real dog... (I'm sure we've talked about this before Bertl)
1113358129 M * ComplexHo They don't update packages on time and they seem to have an attitude of "Wait for our fix".  He says he daren't ask for any of the kernel source for fear of opening a Pandora's box...
1113358202 M * Vudumen yes. and they like to do the installations instead of us... 
1113358209 M * ComplexHo But that's just what he said to me...  Having seen him fixing it up I really would be quite scared to support it professionally
1113358214 M * ComplexHo exactly
1113358233 M * ComplexHo and they format the hard drives in a really bitched up way with _tons_ of inodes
1113358260 M * Vudumen yes but they limit it in vpses too low
1113358261 M * ComplexHo which I'm sure helps but it seems really hacked-up all round
1113358306 M * ComplexHo The interface is it's only real saving grace as far as I can tell - in fact the only way to manage it and not run into serious problems
1113358332 M * ComplexHo And he tells me they are rolling out Viruozzo VPS for Windows ATM ;)
1113358340 M * Vudumen it has some commandline interface-like commands but i didn't tried since it didn't had any usuable docs
1113358350 M * ComplexHo yeah it's really nasty
1113358366 M * Bertl for windows?
1113358369 M * ComplexHo did you get the ifconfig output from # ?
1113358374 M * ComplexHo yeah lol
1113358379 M * ComplexHo honest
1113358389 Q * sebd Remote host closed the connection
1113358477 M * ComplexHo haven't seen it yet but my mate bet a beer for everyone on the windows admin team on them not being able to pull it off properly without using VMs
1113358509 M * ComplexHo and they are trying to collect but we have still to get a proper close-up look at it
1113358528 M * ComplexHo just the thought seems ridiculous somehow
1113358569 M * Vudumen am i right if i say that virtuozzo was asplinux in tha past?
1113358576 M * Vudumen s/tha/the/
1113358579 M * ComplexHo Vudumen, did you see the ifconfig output Virtuozzo gives - did you manage to figure out what was going on?
1113358585 M * ComplexHo yes I believe so
1113358600 M * Vudumen ComplexHo: maybe i saw. now i can't remember
1113358607 M * ComplexHo it's still the same russians running the software, except they are rich now ;)
1113358608 M * Bertl Vudumen: http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-infohide-feat01.diff
1113358621 M * Vudumen Bertl: downloading :)
1113358761 Q * alexx Ping timeout: 480 seconds
1113359116 M * ComplexHo anyway guys time for me to hit the hay... enough hacking for tonight...
1113359153 M * Vudumen good n8 ComplexHo 
1113359164 M * ComplexHo night
1113359188 M * Bertl night ComplexHo!
1113359205 M * ComplexHo night Bertl
1113359288 M * Bertl hmm, the virt_mem looks not so bad actually ...
1113359423 M * Vudumen okkay rebbot with the new kernel
1113359586 M * Vudumen hmm fine :) /proc/self/status seems to be working
1113359593 M * Vudumen let's see some other items in proc
1113359631 M * Vudumen Bertl: uname gives me the right values of the kernel (the faked ones) but /proc/version reveals the true
1113359651 M * Vudumen /proc/swaps shows the real swap
1113359673 M * Vudumen hmm but swap can be hidden
1113359693 M * Bertl hmm, let me check /proc/version to start with something ..
1113359740 M * Bertl /proc/swaps would be hard to virtualize
1113359756 M * Bertl but we could return empty there
1113359757 M * Vudumen yes and it's easier to hide it with vprochide
1113359828 M * Vudumen setattr --admin /proc/swaps
1113359833 M * Bertl yup
1113359838 M * Vudumen is it the correct syntax to hide in vservers?
1113359856 M * Bertl probably --admin --hide
1113359883 M * Vudumen fine :)
1113359893 M * Vudumen do i have to do this after every boot?
1113359910 M * Bertl no, you can just add it to the vprocunhide list
1113359929 M * Bertl (well, or more precisely avoid it there ...)
1113360013 M * Bertl hmm, the version is in fact something which needs 'real' virtualization
1113360067 M * Bertl I can add the kernel side for that in half an hour, but it would need tool side support too ...
1113360084 M * Bertl currently the 'version' contains:
1113360091 M * Bertl "Linux version " UTS_RELEASE " (" LINUX_COMPILE_BY "@"
1113360092 M * Bertl         LINUX_COMPILE_HOST ") (" LINUX_COMPILER ") " UTS_VERSION "\n";
1113360148 M * Bertl so I could do a half assed job by using the LINUX_COMPILE_BY/HOST/COMPILER and mix it with the release and version
1113360169 M * Bertl would that be sufficient?
1113360233 M * Vudumen it would be fine i think :)
1113360256 M * Bertl okay, guess that should be tied to the utsname virtualization
1113360645 Q * Guest392 Remote host closed the connection
1113361064 M * Bertl okay, test compiling banner changes
1113361138 J * berni ~berni@svr01.mucip.net
1113361233 M * Bertl welcome berni!
1113361249 J * alexx ~alexx@82.225.136.176
1113361347 M * Bertl wb alexx!
1113361384 M * Vudumen uhh it's 5:00! :(
1113361490 M * Bertl is that a bad time?
1113361533 M * Vudumen no :) i like this time :) and now i'm on holiday this week (to work on a vpn project) but next week it will be hard to get up early :(
1113361566 M * Bertl that's simple, you 'just' extend it further and further into the day ...
1113361582 M * Bertl you then reach a point where you actually gained a day ;)
1113361582 M * Vudumen yes i used to do this :)
1113361611 M * Vudumen or usually? :)
1113361617 M * Vudumen i should speak more english
1113361703 M * Bertl the channel is a good training ... I wish that native speakers would complain more often about bad grammar/style/vocabulary when they read it ...
1113361724 M * Bertl (and not silently ignore it ;)
1113361730 M * Vudumen yes. the same as developing... hunting bugs is a good thing :)
1113361760 M * Bertl guess you won't like the next patch ...
1113361779 M * Bertl (causes a complete kernel recompile)
1113361790 M * Vudumen no problem :)
1113361900 M * Vudumen hmm. MAKEFLAGS="-j2" make-kpkg [...] doesn't seem to work :/
1113361943 M * Vudumen it wants me to set CONCURRENCY_LEVEL. is it a simple environment variable?
1113361971 M * Bertl no idea about debian's kpkg
1113362112 M * Bertl http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-banner-feat01.diff
1113362163 M * Vudumen i saw other fixes in this dir. are they merged into 1.9.5.x?
1113362178 M * Bertl fixes yes, features partially
1113362186 M * Vudumen ok :)
1113362190 M * Vudumen compiling
1113362281 M * Bertl btw, /proc/locks is also not virtualized ...
1113362298 M * Vudumen fine :)
1113362311 M * Bertl but I'm not sure if that is needed at all
1113362421 M * Vudumen more virtualized is better i think :)
1113362467 M * Bertl in this particular case it's not very hard to implement ... but the testing is probably non-trivial
1113362495 Q * ciphernaut Read error: Connection reset by peer
1113362513 M * Vudumen Bertl: don't you know which driver supports the second ethernet controller on these newer intel server boards? (brandon)
1113362529 M * Vudumen lspic sais: Ethernet controller: Marvell Technology Group Ltd.: Unknown device 4361 (rev 17)
1113362547 M * Vudumen but it cannot be used with tg3 or sk98lin
1113362576 M * Bertl no idea but did you try google?
1113362604 M * Vudumen it's done
1113362610 M * Vudumen not yet. i had otherthings to do :)
1113362675 M * Vudumen intel changed many things. older boards had adaptec scsi controller it has lsi logic (fusion mpt)
1113362691 M * Vudumen older had 2 intel gigabit ethernet controllers it has one intel and one marvell
1113362724 M * Bertl yeah, well, I'm always carefully selecting my boards (with a special focus on the on-board stuff)
1113362729 M * Vudumen and it's very slow to boot up... about two 1,5-1,8 times as the older
1113362770 M * Bertl eh, kernel or mobo?
1113362782 M * Vudumen me too. but for x86 servers i like intel. and these boards supports new xeons (with 800 MHz fsb)
1113362814 M * Vudumen fine ;) it's working
1113362815 M * Vudumen Linux mysql 2.4.22-grsec #1 SMP Sun Oct 5 20:59:21 CET 2003 sparc GNU/Linux
1113362815 M * Vudumen Linux version 2.4.22-grsec (root@agency) (gcc version 3.3.5 (Debian 1:3.3.5-8)) #1 SMP Sun Oct 5 20:59:21 CET 2003
1113362909 M * Bertl I can add the locks virtualization, if you find some way to test it ;)
1113362977 M * Vudumen currently i don't think i can test it :(
1113362994 M * Vudumen a question. why do we have to use the legacy apis? isn't there any tools available to use the new api?
1113363019 M * Bertl you can disable the legacy (not the network legacy) with recent enough tools
1113363047 M * Bertl for the networking the 'legacy' is still the standard ...
1113363049 M * Vudumen i tried with 0.30.205 but it sent me error messages
1113363093 M * Bertl http://vserver.13thfloor.at/Experimental/MDK/util-vserver-0.30.205-3mdk.src.rpm
1113363102 M * Bertl (should work fine with legacy disabled)
1113363133 M * Vudumen tomorrow i look into it i think
1113363140 M * Vudumen now i go to be. i'm a bit tired :(
1113363144 M * Vudumen bed
1113363152 M * Bertl okay, thanks for your help!
1113363166 M * Vudumen no. me say thanks to you :)
1113363175 M * Bertl you're welcome!
1113363180 M * Vudumen have a good night :)
1113363190 M * Bertl and sweet dreams!
1113363194 M * Vudumen :)
1113367774 J * itamarjp lualele@terra-200-225-254-003-dynamic.idial.com.br
1113367777 M * itamarjp heloooooooooooooooooooo
1113367833 M * Bertl welcome itamarjp!
1113367847 M * itamarjp thanks
1113367856 M * itamarjp I need a help to resize a filesystem
1113367864 M * itamarjp Can you help-me ?
1113367874 M * Bertl well, is it linux-vserver related?
1113367909 M * itamarjp Yes, is related to linux
1113367935 M * itamarjp but I am using #$% now, not vserver.
1113367964 M * itamarjp Can i ASK you ?
1113367965 M * Bertl hmm, what is #$% ?
1113367971 M * itamarjp uml
1113367978 M * itamarjp usermodelinux.
1113367996 M * Bertl hmm, and you want to resize an uml disk image?
1113368015 M * itamarjp yes
1113368021 M * itamarjp but I am getting an error.
1113368026 M * itamarjp sample
1113368026 M * itamarjp dd if=/dev/zero of=root_fs.ext3 bs=1 count=1 seek=40G conv=notrunc
1113368044 M * itamarjp  resize2fs -p root_fs.ext3
1113368048 M * itamarjp resize2fs 1.36 (05-Feb-2005)
1113368052 M * itamarjp resize2fs: File too large while trying to determine filesystem size
1113368053 M * itamarjp [root@router conacci]#
1113368077 M * itamarjp .
1113368077 M * itamarjp .
1113368078 M * itamarjp .
1113368079 M * itamarjp '[root@router conacci]# e2fsck -f root_fs.ext3 -y
1113368080 M * itamarjp e2fsck 1.36 (05-Feb-2005)
1113368080 M * itamarjp Pass 1: Checking inodes, blocks, and sizes
1113368080 M * itamarjp Pass 2: Checking directory structure
1113368080 M * itamarjp Pass 3: Checking directory connectivity
1113368080 M * itamarjp Pass 4: Checking reference counts
1113368082 M * itamarjp Pass 5: Checking group summary information
1113368082 M * itamarjp root_fs.ext3: 53583/128000 files (0.2% non-contiguous), 140324/256000 blocks
1113368084 M * itamarjp [root@router conacci]# dd if=/dev/zero of=root_fs.ext3 bs=1 count=1 seek=40G conv=notrunc
1113368084 M * itamarjp 1+0 records in
1113368086 M * itamarjp 1+0 records out
1113368086 M * itamarjp [root@router conacci]# resize2fs -p root_fs.ext3
1113368088 M * itamarjp resize2fs 1.36 (05-Feb-2005)
1113368088 M * itamarjp resize2fs: File too large while trying to determine filesystem size
1113368090 M * itamarjp [root@router conacci]#
1113368090 M * itamarjp [root@router conacci]#
1113368142 M * itamarjp vserver uses a disk image too ?
1113368147 M * Bertl nope
1113368205 M * Bertl but I'd try to use a loop deivce for that and check that the resize2fs tools support large files
1113368224 M * Bertl also it would not be a too bad idea to make the file non-sparse
1113368376 M * itamarjp how to check it
1113368379 M * itamarjp ?
1113368416 M * Bertl a good start would be strace -fF and see if it opens the file with O_LARGEFILE
1113368543 M * itamarjp open("root_fs.ext3", O_RDONLY|O_LARGEFILE) = 4
1113368544 M * itamarjp uname({sys="Linux", node="router.ispbrasil.com.br", ...}) = 0
1113368544 M * itamarjp ioctl(4, BLKGETSIZE64, 0xbfffdba8)      = -1 ENOTTY (Inappropriate ioctl for device)
1113368544 M * itamarjp ioctl(4, BLKGETSIZE, 0xbfffdba4)        = -1 ENOTTY (Inappropriate ioctl for device)
1113368544 M * itamarjp ioctl(4, FDGETPRM, 0xbfffdbb0)          = -1 ENOTTY (Inappropriate ioctl for device)
1113368544 M * itamarjp lseek(4, 1024, SEEK_SET)                = 1024
1113368686 M * itamarjp .
1113368705 M * Bertl looks good for the large file support so far
1113368772 M * itamarjp but not work
1113368800 M * itamarjp there are options for number of open files in linux ?
1113368808 M * itamarjp max open files ?
1113368938 M * Bertl yep, there are .. why? how is that related?
1113368976 M * itamarjp I have many open files openned on this linux box
1113370484 J * erwan_ho ~erwan@lns-vlq-39f-81-56-133-136.adsl.proxad.net
1113370525 M * Bertl night folks ... I'm off to bed ...
1113370526 Q * erwan_ho Remote host closed the connection
1113370534 N * Bertl Bertl_zZ
1113371284 Q * itamarjp Quit: 
1113372022 Q * Jdogg Read error: Connection reset by peer
1113372048 J * Jdogg Jdogg@c-24-1-240-54.hsd1.tx.comcast.net
1113372291 J * MrX freebsd@219.95.237.202
1113376811 Q * mcp Ping timeout: 480 seconds
1113377185 N * Doener|gone Doener
1113377193 M * Doener morning!
1113377332 M * Doener for those interested, i've done a small update to the 2.6 step-by-step guide, added some more explanations, did some improvements and splitted the instructions into smaller sets...
1113379822 J * jsambrook ~jsambrook@80.229.143.166
1113381753 J * Pazzo ~thomas@adsl136-175.aknet.it
1113381754 M * Pazzo moin
1113381761 M * Doener morning Pazzo 
1113381924 J * prae ~prae@134.106-14-84.ripe.coltfrance.com
1113382793 J * DuckMaster ~duckx@195.75.27.158
1113384131 M * Hollow morning
1113384200 M * Pazzo morning Hollow, hi Doener!
1113385274 M * Hollow hm.. can i prohibit processes binding to all interfaces? i have an alias for my vserver and many users starting their own daemons on te host, but they should not be allowed to bind to the alias..
1113385696 M * SiD3WiNDR don't run stuff on host
1113385701 M * SiD3WiNDR that helps
1113385702 M * SiD3WiNDR :)
1113385714 M * Pazzo Hollow: if users on your host are allowed to bind services to unprivileged ports they are per default also allowed to do so on all ip addresses. maybe depending on your setup you can combine their login shell with chbind...
1113385727 M * Pazzo ...but you should ask someone else about details ;-)
1113385742 M * Hollow hm.. good idea.. i will test it
1113385758 M * Hollow shouldn't it be enough to run sshd with chbind?
1113385785 J * gregster ~gregor@62.75.210.70
1113385791 M * gregster hi
1113385819 M * Pazzo Hollow: no idea, maybe. let me know it!
1113386294 M * Hollow Pazzo: seems to work :)
1113386431 Q * jsambrook Ping timeout: 480 seconds
1113386478 M * Pazzo Hollow: thnx! so v_sshd is all you need, right?
1113386507 M * Hollow v_sshd? i just changed my ssh init script
1113386512 J * jsambrook ~jsambrook@aelfric.plus.com
1113386529 M * Hollow start-stop-daemon --start --quiet --pidfile /var/run/sshd.pid --startas /usr/sbin/chbind -- --silent --ip 213.133.98.19/27 --ip 127.0.0.1/8 /usr/sbin/sshd
1113386660 M * Pazzo Hollow: there should be v_* start/stop scripts delivered mit alpha-utils to do exactly this job
1113386669 N * Pazzo PazZzzzooo
1113386673 M * PazZzzzooo (away for lunch)
1113386678 M * PazZzzzooo cya l8er
1113386679 M * Hollow gentoo doesn't ship em
1113386682 M * Hollow cu
1113387413 J * sebd ~sebd@lesdeveloppementsdurables.org
1113389216 J * Jd0gg Jdogg@c-24-1-240-54.hsd1.tx.comcast.net
1113389244 Q * Jdogg Read error: Connection reset by peer
1113389244 N * Jd0gg Jdogg
1113389263 N * PazZzzzooo Pazzo
1113392851 N * Bertl_zZ Bertl
1113392865 M * Bertl morning folks!
1113392969 M * Pazzo morning bertl!
1113392988 M * Hollow hey Bertl 
1113394443 M * Bertl Hollow: hmm why does gentoo remove the v_* scripts?
1113394452 M * Bertl (well, except for the v_sshd script)
1113394522 M * Hollow i did not change the behaviour when i took over maintainership, but the init scripts do not use gentoos init system, the only reason i can think of
1113394544 M * Bertl ah, good point!
1113394563 M * Bertl maybe there should be gentoo specific v_* scripts?
1113394564 M * Hollow though everyone can use the vsyswrapper in /usr/lib/util-vserver
1113394580 M * Hollow thought of that, maybe i convert them
1113394596 M * Bertl that would be great! anyway thanks for the info!
1113394601 M * Hollow sure
1113394643 M * Hollow hopefully i get all stuff in before vserver reaches 2.0
1113394683 M * Bertl well, we have to work on a "2.0" tools release anyway, so you're welcome to join there ...
1113394715 M * Hollow apropos... mind a query?
1113394851 M * Bertl nope
1113395046 M * Pazzo Bertl: what will there be changed / new in 2.0 tools?
1113395100 M * Bertl mostly internal stuff and missing interfaces/features
1113395128 Q * albeiro Ping timeout: 480 seconds
1113395297 J * albeiro albeiro@linux.gentoo.pl
1113396967 P * jsambrook 
1113397961 Q * BWare Read error: Operation timed out
1113398285 J * BWare ~bware@212.26.196.195
1113399267 Q * ntrs Quit: Leaving
1113399699 J * grecea ~grecea@h-195-22-237-74.mdl.net
1113400221 Q * monrad Quit: Leaving
1113400931 J * Jd0gg Jdogg@c-24-1-240-54.hsd1.tx.comcast.net
1113400931 Q * Jdogg Read error: Connection reset by peer
1113400937 N * Jd0gg Jdogg
1113403008 J * ntrs ~ntrs@Dardeene-68.188.50.87.charter-stl.com
1113404120 J * Apollo ~Apollo@caracal.norcomcable.ca
1113404353 P * Pazzo 
1113404511 Q * Jdogg Read error: Connection reset by peer
1113405212 M * Bertl okay folks, leaving now ... back later ...
1113405220 N * Bertl Bertl_oO
1113405860 M * SiD3WiNDR cya Bertl_oO :)
1113408323 Q * prae Quit: Client exiting
1113408976 J * Roger ~Roger@pppoe54-99-luxdsl-251.pt.lu
1113409293 P * Roger No boundaries on the net
1113410186 J * muadda ~muadda@gar31-1-82-234-50-167.fbx.proxad.net
1113410273 Q * MrX Remote host closed the connection
1113411254 J * monrad ~monrad@x1-6-00-09-6b-3f-ee-63.k138.webspeed.dk
1113414924 J * DuckKing ~Duck@dyn-83-157-172-91.ppp.tiscali.fr
1113415352 Q * duckx Ping timeout: 480 seconds
1113415400 J * ntrs_ ntrs@Dardeene-68.188.50.87.charter-stl.com
1113415400 Q * ntrs Read error: Connection reset by peer
1113416008 P * Apollo 
1113417101 J * Borg_Number_One ~borg_numb@p54B8809B.dip0.t-ipconnect.de
1113417106 M * Borg_Number_One Hi.
1113417164 M * Borg_Number_One SS LL AA DD EE NN, I found an original Phoenix-decompressor for IBM compressed BIOS files :) :) :) :)
1113417972 Q * ripman Read error: Connection reset by peer
1113418804 Q * flock Remote host closed the connection
1113419450 J * erwan_ho ~erwan@lns-vlq-39f-81-56-133-136.adsl.proxad.net
1113420783 N * Bertl_oO Bertl
1113420807 M * Bertl B#1 congrats!
1113420813 M * Bertl evening folks! 
1113420819 M * Doener evening Bertl
1113420838 M * Bertl how are you? did you sleep better today?
1113420876 M * Doener feeling just fine, sleep was also fine, thanks
1113420908 J * flock ~restless@l192-117-111-12.broadband.actcom.net.il
1113420995 M * Bertl Doener: excellent ... 
1113421000 M * Bertl evening flock!
1113421017 M * Bertl Doener: so we can get some things done tonight?
1113421028 M * Doener i hope so!
1113421065 M * Bertl ok, great! give me 10 minutes to get my things sorted out ...
1113421252 J * Nik ~Nik@cable-153-130.online.bg
1113421257 M * Nik hi all
1113421382 M * Bertl evening Nik!
1113421394 M * Nik Bertl: :-)
1113421401 M * Borg_Number_One Good evening.
1113422123 J * yarihm ~yarihm@80-218-3-32.dclient.hispeed.ch
1113422175 M * Bertl welcome yarihm!
1113422383 M * Borg_Number_One cu
1113422388 Q * Borg_Number_One Quit: 
1113422569 Q * erwan_ho Remote host closed the connection
1113423032 M * Bertl okay, guess I'm ready ...
1113423063 M * Bertl Doener: http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-infohide-feat01.diff
1113423097 M * Bertl IIRC Vudumen complained that the info_hide is only partial ...
1113423111 Q * Nik Ping timeout: 480 seconds
1113423116 M * Bertl (this fixes it, and was at least tested by him)
1113423154 M * Bertl I also 'included' your x25 cleanup, looks fine to me (http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-x25-fix03.diff)
1113423171 J * Nik ~Nik@cable-153-130.online.bg
1113423202 M * Bertl and I cleaned up the proc_mounts stuff (http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-proc_mounts_fix04.diff ;)
1113423233 M * Bertl in addition I virtualized the banner information (with the utsname parts)
1113423238 M * Bertl http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-banner-feat01.diff
1113423320 M * Doener hm, i only updated my 2.6 guide... guess i got to catch up :)
1113423337 M * Bertl s/only/just/
1113423432 M * Bertl ah, and here is an inode cleanup (commands are separated out)
1113423433 M * Bertl http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-inodes-clean01.diff
1113423539 J * nayco ~nayco@lns-vlq-47-nan-82-252-247-215.adsl.proxad.net
1113423553 M * Doener yep, patches seem to be fine
1113423554 M * yarihm yo Bertl 
1113423559 M * nayco 'llo !
1113423971 Q * Nik Ping timeout: 480 seconds
1113424159 Q * flock Remote host closed the connection
1113424160 M * Bertl welcome nayco!
1113424193 M * Bertl Doener: okay, enrico requested that we extend the vkill (with pid=0) to the init process ...
1113424236 M * Bertl and I really want to make the vshelper synchroneous, any comments to that?
1113424276 M * Doener hm, what does it do with pid 0? IIRC i've seen pid -1 to be used to kill the whole context (kernel side)... 
1113424380 M * Bertl hmm, maybe I'm confusing 0 and 1 here ...
1113424390 M * Bertl 0 and -1
1113424411 M * Bertl give me a second to look at the code ...
1113424438 M * Doener both do the same :)
1113424461 M * Doener guess i just remembered -1 because it looked a little weird on first sight
1113424473 M * Doener s/just/only/
1113424488 M * Bertl okay, so what if we make the -1 kill all + init
1113424500 M * Bertl (i.e. use the kernel type signal (unblockable)
1113424506 M * Bertl and leave the 0 as it is?
1113424571 M * Bertl btw, it seems to me that there is at least a line missing in my tree
1113424754 M * Bertl http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-vkill-fix01.diff
1113424759 M * Bertl yup that one was missing ...
1113424875 M * Doener hm, "kill -1" is "kill all but myself and init"... so maybe we should do it like that?
1113424892 M * Doener i.e. keep -1 behaviour and change 0 behaviour
1113424894 M * Bertl ah, and 0 is killall?
1113424906 M * Doener no, 0 seems to have no special meaning
1113424940 M * Bertl okay, then we do it this way ...
1113424983 M * Doener oh wait... 0 is every process in the current process group... 
1113425015 M * Bertl well, that matches, no?
1113425019 M * Doener yep, it does
1113425259 M * Bertl so what about this ontop of feat01 without fix01 ? http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-vkill-fix02.diff
1113425405 M * Doener yep, looks good
1113425632 M * Bertl okay, I'll fix the vshelper (i.e. make it synchronus) could you prepare some test setup to verify that this is working with 'slightly' modified tools?
1113425682 M * Bertl btw, probably it would be also a good idea to add the start/stop state vshelper calls, no?
1113425718 M * Doener guess so... we'll need them anyways
1113426289 M * Doener got a test setup ready (in qemu)...
1113426302 M * Bertl 2fast4me ;)
1113426330 M * Doener well, it was ready before i started, just checked that the current setup is fine
1113426368 M * Bertl do we need the RESTART2 command btw?
1113426391 M * Bertl I really don#t see a good use for it .. so I'd like to remove it ...
1113426405 M * Bertl LINUX_REBOOT_CMD_RESTART2
1113426430 M * Doener no idea... at the kernel side it looked like RESTART1 and RESTART2 are basically the same atm... but i remember some discussion about that on the mailing list...
1113426455 M * Bertl linux-vserver ml or kernel ml?
1113426459 M * Doener lv
1113426566 M * Bertl does the vshelper handle it somehow?
1113426619 M * Doener simply does nothing
1113426760 M * Bertl okay, so we remove it ;)
1113427102 M * Bertl http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-vshelper-clean01.diff
1113427115 Q * click Read error: Connection reset by peer
1113427117 M * Bertl okay, doesn't do the start/stop yet .. but I guess you can start testing with that one
1113427209 M * Doener k
1113428048 M * Doener hm, is 1.9.5.x fine? if not, what should i get ontop of that?
1113428137 Q * monrad Quit: Leaving
1113428177 M * Bertl sec, let me upload a 'current' version ;)
1113428185 M * Doener ok *cancels build*
1113428199 M * Doener i should learn to ask first, then build ;)
1113428365 M * Bertl http://vserver.13thfloor.at/Experimental/patch-2.6.11.7-vs1.9.5.x5.diff.bz2
1113428479 M * Doener hm, we still send restart2 as default
1113428497 M * Bertl hmm, right ...
1113428508 M * Bertl I'll remove that in the next run
1113428741 Q * albeiro Ping timeout: 480 seconds
1113428857 J * albeiro albeiro@linux.gentoo.pl
1113429087 J * lost_eps ~lost_eps@c-24-98-52-192.hsd1.ga.comcast.net
1113429264 M * Bertl welcome lost_eps!
1113429298 M * lost_eps hi!
1113429376 M * lost_eps running FC3, 2.6.11.5 vserver-1.9.5 - SElinux disabled
1113429386 M * lost_eps everytime I stop a vserver I get:
1113429388 M * lost_eps Shutting down kernel logger:                               [FAILED]
1113429388 M * lost_eps Shutting down system logger:                               [  OK  ]
1113429388 M * lost_eps Starting killall:                                          [  OK  ]
1113429407 M * lost_eps is that kernel logger failed normal?
1113429414 M * Bertl and you wonder about the klogd?
1113429422 M * Doener aw shit... i got bad head ache... can't continue...
1113429423 M * lost_eps correct
1113429473 M * Bertl well, you can 'upgrade' to a more recent kernel, or add the virtual syslog patch or just remove the klogd from the script
1113429492 M * Bertl Doener: hmm, np
1113429552 M * Bertl lost_eps: http://vserver.13thfloor.at/Experimental/FOR-1.9.6/delta-syslog-feat01.diff
1113429566 M * Doener hm, init-based vserver died on reboot... is the init protection included in x5?
1113429580 M * lost_eps great, thank you
1113429593 M * Bertl Doener: good question, but I'd say yes ...
1113429637 M * Doener hm, yes it is... maybe i just didn't wait long enough, for the vserver to be restarted... i'll know tomorrow...
1113429641 M * Doener good night folks...
1113429645 N * Doener Doener_zZz
1113429648 M * Bertl night Doener_zZz!
1113429667 M * lost_eps will patch-2.6.11.5-vs1.9.5.diff patch cleanly with higher numbered kernels?
1113429687 M * Bertl http://vserver.13thfloor.at/Experimental/patch-2.6.11.7-vs1.9.5.x.diff.bz2
1113429699 M * Bertl (or are you looking at 2.6.12-rc?
1113429702 M * Bertl +)
1113429723 M * lost_eps no, not looking at 12-rc
1113429734 M * lost_eps thanks
1113429747 M * Bertl you're welcome
1113429747 M * Bertl btw, that includes the syslog feature
1113429775 M * lost_eps sweet, two birds with one stone
1113430402 M * nayco Bertl: Have you seen my answer on the ML ?
1113430419 M * Bertl not yet ...
1113430423 M * nayco k
1113430438 A * Bertl was working his way through his mailbox ... 
1113430456 M * Bertl well, actually I am still ;)
1113430477 M * nayco er... It was a simple question: How can I get the debug logs, since I recompiled my kernel with debug support ?
1113430520 M * Bertl ah, you simply enable debugging via /proc/sys/vserver/debug_*
1113430541 M * Bertl depending on what you want to debug, you write different values there
1113430561 M * nayco where are they documented ?
1113430574 M * Bertl in the source ;)
1113430576 M * nayco funny: I tried to tail those files ;-)
1113430591 M * nayco "Use the source", they said :)
1113430596 M * lost_eps Bertl I'm still new to linux-vserver was curious if it was documented anywhere that I missed about the vserver networking setup?
1113430627 M * lost_eps you do not see the IP address using ifconfig when entering a vserver as you do with alternatives such as virtuozzo or user-mode-linux
1113430646 M * Bertl nayco: basically they are bit coded so enabling more bits (e.g. writing 255 there) will enable more debug info
1113430675 M * nayco Bertl: So, writing 255 may enable full logging, right ?
1113430679 M * Bertl lost_eps: that's because by default the util-vserver tools use the 'ip' tool (from iproute2) to configure the ip addresses
1113430697 M * Bertl and those do not show up with ifconfig unless they are named aliases
1113430715 M * Bertl nayco: yes, if you write it to each of them, it will enable all debugging ...
1113430724 M * lost_eps I see
1113430757 M * lost_eps this may be good to add to a howto if it has not been done already to save confusion to newbies
1113430758 M * Bertl lost_eps: so you _can_ see the ips pretty fine with 'ip' inside or you can specify --interface name=<ip>/<prefix>
1113430782 M * lost_eps or FAQ
1113430802 M * nayco Ok, i'm reading the source now, it seems that the debug stuff is at the very beginning... Cool.
1113430802 M * Bertl yes, we have a FAQ and IIRC it is documented somewhere ... 
1113430809 M * lost_eps good to know, thank you!
1113430841 M * Bertl you're welcome!
1113430862 M * Bertl ah, and btw, feel free to enhance the documentation, it's a wiki!
1113430894 M * lost_eps as I learn more, I'll do just that ;)
1113430993 M * nayco lost_eps: example for building a vserver with an alias:\n vserver test1 build -m skeleton --hostname test1.yourdomain.fr --interface test1=eth0:172.16.210.30 (If i'm not mistaken)
1113431015 M * nayco ...forget the \n ;)
1113431031 M * lost_eps haha, nayco thanks, saving that now
1113431120 M * lost_eps here's a question......is there such thing as Copy On Write support with linux-vserver? so your not using as much disk space to create a new vserver ?
1113431149 M * Bertl yes and no
1113431163 M * Bertl we do not do the cow stuff at the block layer or so ...
1113431176 M * Bertl instead we use something called unification
1113431189 M * Bertl that's basically hard links with an extra protection
1113431198 M * Bertl you can not modify them, but safely remove them
1113431231 M * Bertl (which basically all package systems do quite fine)
1113431240 M * lost_eps ahh, did not realize that
1113431268 M * Bertl there is also a cow-link patch in development/testing
1113431293 M * Bertl it does break the links on the first write ... but as I said, it's not really required
1113431609 M * nayco lost_eps: I've been writing a small "getting started" guide... It's a real draft for now, the titles are in french, but the contents in english... ;) Well, take a look, there may be something to learn here : http://nayco.free.fr/wiki/doku.php?id=linux-vserver
1113431609 M * Bertl ah, and almost forgot, it has another advantage ...
1113431641 M * Bertl hard linked files get only buffered/mapped once, so you also save caches and buffers
1113431710 Q * yarihm Quit: Leaving
1113431840 M * lost_eps are you referring to unification or the COW patch?
1113431841 Q * muadda Quit: Leaving
1113431848 M * lost_eps nayco thanks, I'll check it out
1113431854 M * Bertl the unification
1113431991 M * lost_eps sorry if this is a dumb question, I'm sure I'm staring at it, but is there a page that covers how to setup unification?
1113432183 M * nayco k guys, good night (I have to play with debugging tomorrow ;-)) !
1113432188 M * Bertl night nayco!
1113432203 Q * nayco Quit: Bonne nuit !
1113432208 M * Bertl lost_eps: there is a tool (vunify and recently vhashify) which does that
1113432270 M * lost_eps cool, I'll play with that
1113432293 M * lost_eps I'm getting some dinner thanks for all of your help guys! you've been great help
1113432324 M * Bertl you're welcome! enyou your meal!
1113432467 M * lost_eps later
1113432469 Q * lost_eps Quit: Leaving
1113434685 J * monrad ~monrad@213083190130.sonofon.dk
1113434714 M * Bertl evening monrad!
1113435775 M * monrad evening