1107650241 Q * boklm Remote host closed the connection
1107650318 J * boklm ~boklm@woper.mars-attacks.org
1107650330 M * Bertl wb boklm!
1107651283 M * Doener Bertl: http://doener.homeip.net/doener/vserver/diff-2.6.11-rc3-vs1.9.4-rc4-fix-tcpdiag_dump.diff
1107651283 M * Doener going to test that one now...
1107652075 M * Doener Bertl: yep, works :)
1107652089 M * Bertl great!
1107652367 M * Doener hm, not perfect yet... time-wait got broken
1107652899 M * Bertl okay, I'm off to bed for tonight ...
1107652923 M * Bertl good luck, if necessary, I'll help tomorrow ;)
1107652960 M * Doener hehe, guess i copied that bug from you ;)
1107652999 M * Doener /proc/net/tcp also doesn't listen TIMEWAIT sockets on 1.9.4-rc3 ;)
1107653029 M * Bertl hmm, IIRC I fixed something in ngnet ...
1107653201 M * Bertl k, off for today ...
1107653206 N * Bertl Bertl_zZ
1107654190 Q * _are_ Ping timeout: 480 seconds
1107654849 J * _are_ ~are@dsl-084-056-138-019.arcor-ip.net
1107656212 J * switch ~switch@61.206.115.5.user.ad.il24.net
1107657612 Q * linux-addict Remote host closed the connection
1107657693 J * linux-addict ~noname@145.116.6.75
1107658173 Q * linux-addict Remote host closed the connection
1107658242 J * linux-addict ~noname@145.116.6.75
1107664764 J * lilo ~lilo@lilo.usercloak.oftc.net
1107666741 J * agent ~smith@220.224.22.218
1107666802 P * agent 
1107682228 Q * Loki|muh Remote host closed the connection
1107682258 J * Loki|muh loki@213.133.125.66
1107686330 Q * mcp Ping timeout: 480 seconds
1107687586 J * halorgium tim@halorgium.net
1107691579 J * prae ~prae@62.212.100.153
1107694535 Q * _are_ Quit: Disconnecting
1107697553 J * DuckKing ~Duck@dyn-83-157-198-58.ppp.tiscali.fr
1107697980 Q * duckx Ping timeout: 480 seconds
1107697995 Q * DuckMaster Ping timeout: 480 seconds
1107698012 J * duckx ~Duck@dyn-83-157-198-58.ppp.tiscali.fr
1107699780 Q * serving Ping timeout: 480 seconds
1107703124 Q * linux-addict Remote host closed the connection
1107703191 N * Bertl_zZ Bertl
1107703197 M * Bertl morning folks!
1107703387 M * Bertl morning prae! 
1107704411 M * prae Hi Bertl 
1107704920 J * linux-addict ~noname@145.116.6.75
1107704985 M * Bertl welcome linux-addict!
1107705071 M * linux-addict helo Bertl 
1107705082 M * linux-addict I'm allready a vserver user
1107705092 M * linux-addict u should say welcome back ;)
1107705117 M * linux-addict I'm waiting for a stable release for 2.6 kernel
1107705118 M * Bertl welcome back linux-addict!
1107705150 J * shuri ~shuri@219.200-ppp.3menatwork.com
1107705161 M * Bertl well, waiting for it is the wrong method, start _working_ on it, right now ;)
1107705170 M * Bertl welcome shuri!
1107705224 M * shuri hello Bertl 
1107705235 M * shuri i am back from space
1107705279 M * Bertl hmm, how was it?
1107705295 M * shuri fun :)
1107705357 M * Bertl and I thought it would be .. hum, empty ;)
1107705371 M * linux-addict it is not empty
1107705374 M * linux-addict but full with energy
1107705459 M * shuri hehe
1107705473 M * Bertl prae: so what are your plans at Mandrake? any details?
1107705897 M * Bertl shuri: already tested new ngnet?
1107706012 M * shuri net yet
1107706043 M * shuri well i started but the patch for iptables was not working
1107706069 M * Bertl there are two aptches available one for 1.2.9 and one for 1.2.11 iirc
1107706096 M * shuri ok
1107706789 M * Bertl okay, off for dinner now, back in 40m
1107706804 N * Bertl Bertl_oO
1107708649 N * Bertl_oO Bertl
1107708668 M * Bertl I'm back
1107708759 J * _are_ ~are@dsl-084-056-138-019.arcor-ip.net
1107708908 M * Bertl welcome _are_!
1107709716 J * shuri_ ~shuri@dsl.speedline209.226.electronicbox.net
1107709822 Q * shuri Read error: Operation timed out
1107709988 Q * shuri_ Quit: 
1107710442 Q * linux-addict Remote host closed the connection
1107710447 M * _are_ hi
1107710532 M * _are_ not my day today, managed and stop a vserver with NET_ADMIn cap. and it took down the mains ervers network, despite 'nodev' file and missing both,network configuration and /etc/init.d/networking
1107710847 M * Doener _are_: guess you hit "reboot -i" ;)
1107711007 M * Bertl hmm ...
1107711909 J * nayco ~nayco@lns-vlq-47-nan-82-252-237-136.adsl.proxad.net
1107712000 M * _are_ Doener: I hit 'vserver vservername stop'
1107712027 M * Doener and your runlevel 6 probably includes the reboot script, which calls "reboot -i"
1107712042 M * _are_ might be
1107712052 M * _are_ so another think to kill...
1107712063 M * _are_ it writes down something about rebooting with 'stop' indeed
1107712072 M * _are_ to late now, box is dead :->
1107712161 M * Doener yeah, but you know what to fix ;)
1107712237 Q * nayco Quit: Bonne nuit !
1107712381 M * _are_ this is why I asked
1107712421 M * _are_ i guess i should write down a list of things better to change with a fresh install with vserver and debian
1107712447 M * Bertl _are_: any reason for CAP_NET_ADMIN?
1107712474 M * _are_ if I gave it: definitly
1107712511 M * Bertl I mean, why did you add it? any specific reasons for that?
1107712529 M * _are_ yes, don't know which server it had been.
1107712534 M * _are_ dhcp
1107712603 M * _are_ i can check when the box is up again
1107712637 M * _are_ one box on that vserver is the qemu-box
1107712646 M * _are_ that once has NET_ADMIN now as well
1107713038 M * Bertl ah, i.c. so, it was accidentially then?
1107713105 M * Bertl Doener: did you finish the patch regarding netlink issues?
1107713208 M * _are_ ofc accidentally
1107713220 M * Doener i fixed the vx_check for timewait sockets, but they still don't show. guess that's a remaining bug that also affects your /proc/net/tcp check for timewait.
1107713229 M * _are_ had trouble unmounting some directory, so stopped the vservers
1107713260 M * Bertl Doener: okay, you ahve a test to check this (for coorect behaviour?)
1107713331 M * Doener on my box i simply start fetchmail, which causes some timewait sockets to idle around for some time. then with "chcontext --ctx 1 netstat -tna" you see the tw sockets, but not with "netstat -tna", same for "ss -a"
1107713387 M * Bertl ah, getchmail on the host, right?
1107713411 M * Bertl s/get/fet/
1107713416 M * Doener yep
1107713427 M * Bertl k, where is the patch so far?
1107713444 M * Doener but same applies to the vserver, no tw sockets anywhere but in ctx 1
1107713503 M * Doener http://doener.homeip.net/doener/vserver/diff-2.6.11-rc3-vs1.9.4-rc4-fix-tcpdiag_dump-2.diff
1107713517 M * Bertl tx
1107713613 M * Bertl hey your vserver dir starts filling up ;)
1107713665 M * Doener yeah, once again, but this time without too much crap ;)
1107713736 M * Bertl okay, I'm now working with:
1107713765 M * Bertl 1.9.4-rc4 + delta-64bit-fix01.diff, delta-copy_mm-fix01.diff, delta-schedprio-fix02.diff ...
1107713775 M * Bertl (just FYI)
1107714027 M * Doener i'm off to my gf... will try to ger her laptop to come back here soon :)
1107714032 N * Doener Doener|gone
1107714038 M * Bertl cya
1107714186 J * nish ~nish@220.224.39.253
1107714239 M * nish evening all :)
1107714317 M * Bertl evening!
1107715193 M * Bertl nish: no questions today?
1107716508 M * prae Bertl: we have a meeting tomorrow for roadmap
1107716570 M * prae sorry, i must go now, see'you tomorrow
1107716572 M * prae bye !
1107716582 M * Bertl cya
1107716590 Q * prae Quit: Pwet
1107716621 Q * flock Ping timeout: 480 seconds
1107716854 M * nish not yet :)
1107717660 J * flock ~restless@l192-117-111-12.broadband.actcom.net.il
1107718015 J * doener ~doener@pD95883E3.dip.t-dialin.net
1107718026 M * doener re
1107718064 M * Bertl hmm, your gf seems very dominant ... *G*
1107718083 M * doener hu?
1107718096 M * Bertl because the lowercase 'doener' ;)
1107718124 M * doener ah ;)
1107718167 A * nish runaway from girls
1107718381 J * Nik ~Nik@cable-153-130.online.bg
1107718386 M * Nik hi all
1107718417 M * Bertl welcome Nik!
1107718422 M * Nik Bertl :-)
1107718954 J * linux-addict ~noname@145.116.6.75
1107718971 M * Bertl wb linux-addict!
1107719049 M * linux-addict ty Bertl 
1107719836 M * doener Bertl: i guess the bug is that inet_realease
1107719840 M * doener oops
1107719872 M * Bertl hmm?
1107719883 M * doener the bug is that inet_release already sets sk_xid to -1, tcp_time_wait is called _after_ that one
1107719924 M * doener so we get tw_xid == -1 thus it is only visible in ctx 1
1107719937 M * Bertl hmm .. k, well, that changed recently (maybe just in ngnet)
1107719946 A * Bertl checks with the sources ...
1107720064 J * chrish01 ~chrish01@69.90.131.10
1107720071 M * Bertl welcome chrish01!
1107720081 M * chrish01 hey bertl :)
1107720085 M * Bertl doener: hmm, yep, it's in ngnet ...
1107720108 M * Bertl maybe we should rip it out/add it to mainline?
1107720149 M * chrish01 Bertl: is vserver work well with SELinux?
1107720154 M * chrish01 s/is/does
1107720215 M * doener Bertl: i'd say so... the tw bug shouldbe fixed after all ;) are you going to prepare a patch? i'll focus on my gf now ;)
1107720329 M * doener off now, back later (probably;)
1107720332 Q * doener Quit: leaving
1107720406 M * Bertl chrish01: it's mainly orthogonal ...
1107720417 M * chrish01 ah gotcha
1107720448 M * Bertl of corse you can hinder vserver startup with abad config ...
1107720478 M * chrish01 i got a M4 wizard we just picked up. has lots of experience writing selinux configs
1107722180 M * ntrs_ Bertl, is there a fixed new patch? Basically rc4 + the sched_prio fix?
1107722262 M * Bertl not yet, but rc5 will be out in an hour or so ...
1107722472 M * ntrs_ Changes between rc4 and rc5?
1107722623 M * Bertl fixes in the refcounting basically
1107722641 M * ntrs_ plus the sched_prio I suppose.
1107722645 M * Bertl yes
1107722672 M * ntrs_ Is there any way I could apply the sched_prio fix on rc4? I need one kernel build quckly.
1107722723 M * Bertl the patch is against rc4 ...
1107722738 M * ntrs_ Where is the patch?
1107722745 M * Bertl http://vserver.13thfloor.at/Experimental/RC-1.9.4/delta-schedprio-fix02.diff
1107722766 M * ntrs_ Ok, that will work I guess.
1107722769 M * ntrs_ Thanks Bertl 
1107722775 M * Bertl you're welcome!
1107722797 M * Bertl btw, you can also add the copy_mm fix ...
1107722809 M * Bertl http://vserver.13thfloor.at/Experimental/RC-1.9.4/delta-copy_mm-fix01.diff
1107722816 M * ntrs_ What is that?
1107722858 M * Bertl a bugfix for the mm handling
1107722867 M * ntrs_ mm?
1107722871 M * ntrs_ memory manager?
1107722881 M * Bertl yes
1107722884 M * ntrs_ How is that bug manifested?
1107722933 M * Bertl we don't know _if_ it manifested at all, it might be related to the 'hanging' contexts .. but it was discovered by doener by source code auditing ...
1107722944 M * ntrs_ hmm. ok.
1107724117 J * Hollow ~bene@home.xnull.de
1107724121 M * Hollow heya!
1107724219 M * Hollow Bertl: any plans when 1.9.4 will be released?
1107724264 M * Bertl soon I'd say, and hello Hollow!
1107724278 M * Bertl we are preparing rc5 now ... with some fixes
1107724374 M * ntrs_ Bertl, I actually had rc3. What are the changes between rc3 and rc4?
1107724425 M * Hollow Bertl: sounds good, i just talked to a baselayout guy to get vserver integrated to gentoos baselayout soon
1107724513 M * ntrs_ Bertl, is there a rc3-rc4 delta?
1107724563 Q * nish Read error: Connection reset by peer
1107724657 M * Bertl ntrs_: http://vserver.13thfloor.at/Experimental/RC-1.9.4/delta-2.6.11-rc3-vs1.9.4-rc3-rc4.diff
1107724700 M * Bertl Hollow: looks like distros are getting interested in linux-vserver ... debian, mandrake, gentoo ...
1107724760 M * ntrs_ Thanks
1107724770 M * Hollow Bertl: yay! i hope to get the patches into baselayout before the 2005.0 release
1107724805 M * Bertl sounds good! who will maintain the stuff?
1107724819 A * Bertl is just curious ...
1107724827 M * Hollow me :)
1107724843 M * Bertl okay, so you will keep in touch then, great! ;)
1107724849 M * Hollow for sure!
1107724895 M * Hollow setup my vserver testing machine yesterday with 2.6.10-vs1.9.4-rc4, running fine for now, though it's not really stressed :P
1107724922 M * Hollow i'll try to catch up with ngnet asap
1107724969 M * Bertl k, but keep in mind, ngnet is still highly experimental
1107725050 M * Hollow yep, but i'm really curious about it, simply didn't have time lately :(
1107725166 J * monrad ~monrad@213083190130.sonofon.dk
1107725201 M * Bertl welcome monrad!
1107725356 M * monrad hi
1107725460 M * ntrs_ Bertl, do I need to apply delta-activate-fix01.diff
1107725463 M * ntrs_ ?
1107725499 M * Bertl probably (if you are with rc3)
1107725511 M * ntrs_ No, I just downloaded rc4
1107725577 M * Bertl try with --dry-run, if it complains about reversed stuff, it's probably included
1107725587 M * ntrs_ ok
1107725641 M * ntrs_ yes it is probably included.
1107725642 M * ntrs_ Thanks
1107725647 M * Bertl np
1107727350 Q * monrad Read error: Operation timed out
1107728115 J * monrad ~monrad@213083190130.sonofon.dk
1107728144 Q * linux-addict Remote host closed the connection
1107728541 Q * Nik Quit: Hybernating my bran.....
1107728763 M * pusling hmm.. is there a vnetstat or something like that ?
1107729034 M * Bertl chcontext --ctx 1 netstat
1107729138 M * pusling thx
1107730028 N * _are_ are|afk
1107730587 N * Doener|gone Doener
1107730594 M * Doener re
1107730608 M * Bertl wb Doener!
1107730625 M * Bertl did you abandon your gf that soon?
1107730683 M * Doener she got to get to work in 8 hours and needs her sleep. i got up to late to go to sleep now, so i'm back at home ;)
1107730734 M * Bertl sounds too familiar ;)
1107730765 M * Bertl okay, I fail to get a tcp socket in TIME_WAIT state ...
1107730789 M * Bertl I got a CLOSE_WAIT, LISTEN and ESTABLISHED ...
1107730897 M * Doener hmm... let's see if i can come up with something to force some tw sockets... i guess my connection's latency or so is too high, so that i can see them
1107730933 M * Bertl something which can be reproduced with nc/iperf/telnet would be great ...
1107731324 M * Bertl Doener: see #kernelnewbies ...
1107731350 M * Doener great!
1107731745 M * Bertl hmm, now my question is, what exactly does you patch fix?
1107731838 M * Doener it hides sockets which do not belong to the context, when using "ss" inside a vserver, v2 only fixes the tw check in tcpdiag, but not yet the bug that causes tw socket to only show up in ctx 1
1107731860 M * Bertl ah, okay ... I have the 'other' patch then ... sec
1107731874 J * Rusty Rusty@3e44b2f9.adsl.enternet.hu
1107731882 M * Rusty hello guys
1107731884 M * Bertl welcome rusty!
1107731942 M * Doener v1 had an additional tw bug ;)
1107731945 A * halorgium is pondering the ability to hide routes (ip route list)  on the host box from any children
1107731982 M * Bertl hmm, on what basis?
1107732056 M * Bertl Doener: http://vserver.13thfloor.at/Experimental/RC-1.9.4/delta-sockref-fix02.diff
1107732070 M * Bertl (this together with your patch should make those things behave)
1107732096 Q * Rusty Quit: 
1107732131 M * Bertl halorgium: we could hide all routing entries not related to visible interfaces and used ip addresses for example?
1107732131 J * Rusty Rusty@3e44b2f9.adsl.enternet.hu
1107732131 Q * Rusty Quit: 
1107732164 M * halorgium Bertl: yes ... that is what i mean
1107732186 M * halorgium i have 2 interfaces on my host box ... and only one of them is "forwarded" through
1107732188 J * Rusty Rusty@62.68.178.249
1107732193 M * Bertl any algorithm for that? i.e. something which can be checked from the output?
1107732208 M * Doener hm, i just noticed that gateways are hidden in a vserver when using "route", but not using "ip r"
1107732212 M * halorgium do you want output?
1107732222 M * halorgium yes
1107732264 M * halorgium also inet6 addresses are exposed in `ifconfig'
1107732275 M * halorgium but not exposed in ip -6 addr
1107732320 M * Bertl well, the main question is, do we really want to fix this or are we better off concentrating on ngnet ...
1107732331 M * Doener votes for ngnet :)
1107732340 M * halorgium Bertl: i'm more putting it out there :D
1107732356 M * Bertl halorgium: appreciated!
1107732404 M * Doener guess i finally have to catch up with ngnet code soon ...
1107732524 M * Doener Bertl: hm, no removal of set_xx_info for inet sockets in that patch... 
1107732544 M * Doener nvm, found it
1107732576 A * Doener hates that diff also uses labels to mark hunks...
1107732620 M * Bertl yep, that's a little confusing .. 
1107732651 M * Doener looks good then
1107732657 M * Bertl but I got me a new tool .. well an old tool, but a new (patched) version ...
1107732671 M * Bertl of the 'patch' utility ...
1107732703 M * Bertl patch --help
1107732708 M * Bertl .. bla bla ...
1107732713 M * Bertl   --unified-reject-files  Create unified reject files.
1107732715 M * Bertl   --global-reject-file=file  Put all rejects into one file.
1107732720 M * Bertl .. bla bla ...
1107732727 M * Doener nice
1107732783 M * Rusty is ngnet member of the stable vserver 1.2.10?
1107732785 J * mcp ~hightower@www.c-tera.de
1107732785 M * daniel_hozac very nice. URL? :)
1107732814 M * Doener Rusty: no, it's still highly experimental and will probably never go into the 1.2.x branch
1107732837 M * Bertl welcome mcp!
1107732852 M * Rusty i see
1107732856 M * Bertl daniel_hozac: for the patch?
1107732863 M * daniel_hozac yeah.
1107732937 M * Bertl http://vserver.13thfloor.at/Stuff/MANDRAKE/patch-2.5.9-2mdk.src.rpm
1107732968 M * daniel_hozac thanks.
1107732981 M * Bertl my pleasure!
1107733327 Q * chrish01 Quit: chrish01
1107733685 Q * ndim Ping timeout: 480 seconds
1107733722 M * Bertl Doener: looks good so far, any complaints from your side? andy test results yet?
1107733793 M * Doener no complaints, no test results yet, but in some minutes there will be some
1107733819 M * Bertl great, thanks a lot!
1107733918 J * ndim U2FsdGVkX1@helena.bawue.de
1107733946 M * Bertl welcome ndim!