1146182455 Q * ntrs Ping timeout: 480 seconds
1146182561 M * Bertl daniel_hozac: ah, now I remember what I wanted to ask you regarding scripts
1146182583 M * daniel_hozac oh?
1146182614 M * Bertl it would be nice to have a magic script which extracts all vserver specific defines (maybe also static inlined functions) and checks if they are used/referenced at all
1146182626 J * DEac- ~deac@xdsl-81-173-139-142.netcologne.de
1146182637 M * Bertl it could be quite simple, i.e. no guarantee for correctness
1146182641 J * ntrs ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com
1146182645 M * daniel_hozac hehe.
1146182658 M * Bertl just to get an idea what is never ever used
1146182688 M * daniel_hozac i guess using cscope, that sort of script shouldn't be too hard.
1146182701 M * Bertl yeah, something like that was on my mind ...
1146183088 M * daniel_hozac hmpf. *** stack smashing detected ***: cscope terminated
1146183122 Q * ntrs Ping timeout: 480 seconds
1146183363 J * ntrs ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com
1146183616 Q * jkl jupiter.oftc.net neutron.oftc.net
1146183616 Q * mountie jupiter.oftc.net neutron.oftc.net
1146183616 Q * VAndreas jupiter.oftc.net neutron.oftc.net
1146183616 Q * shedi jupiter.oftc.net neutron.oftc.net
1146183684 J * jkl eric@c-71-56-216-223.hsd1.co.comcast.net
1146183684 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com
1146183684 J * VAndreas ~Hossa@212.110.98.7
1146183684 J * shedi ~siggi@inferno.lhi.is
1146183786 M * Bertl daniel_hozac, doener: do we have an idea what flags would be a good default for legacy?
1146183793 M * Bertl eyck: ping!
1146183965 Q * ag- Ping timeout: 480 seconds
1146183965 Q * nebuchadnezzar Ping timeout: 480 seconds
1146183974 J * nebuchadnezzar ~nebu@zion.asgardr.info
1146184019 M * daniel_hozac hide_netif, info_init, virt_uptime seems to be the default for new guests.
1146184068 M * doener hide_netif, hide_mount, virt_uptime, maybe info_hide?
1146184097 M * doener daniel_hozac: info_init is the blend-through flag? wasn't there some option for that in the legacy config as well?
1146184104 J * ag- ag@caladan.roxor.cx
1146184132 M * daniel_hozac hmm, right, there is.
1146184164 M * Bertl yes, but it had different meaning
1146184172 M * Bertl basically the fake vs real init
1146184180 M * Bertl (IIRC)
1146184216 M * daniel_hozac is hide_mount used?
1146184241 M * doener dunno, just picked some of the flags defined in the kernel headers ;)
1146184274 M * doener virt_load might be nice, but that would be a change in behaviour IIRC
1146184348 M * daniel_hozac ah, hide_mount is for when you just enter the namespace and don't chroot?
1146184410 M * daniel_hozac hmm, no, hide_mount hides all of them?
1146184414 M * doener hm, seems to hide all mounts?
1146184434 M * daniel_hozac probably not that good of a default then.
1146184440 M * doener ok, then we should probably not make that the default, I thought that was another condition for hiding, like HIDE_NETIF
1146184452 M * daniel_hozac me too.
1146184951 M * Bertl okay, so hide_netif, virt_uptime, info_init sounds good?
1146184973 M * doener what was the purpose of info_hide?
1146184974 M * daniel_hozac yep.
1146185111 M * daniel_hozac hiding vserver information in /proc/<pid>?
1146185142 M * Bertl ah, that might make sense too, indeed
1146185156 M * daniel_hozac except nid and xid in /proc/<pid>/[nv]info.
1146185247 M * doener i don't remember the original behaviour for that... could that break anything useful?
1146185265 M * daniel_hozac testme?
1146185286 M * Bertl I should have a guest running somewhere on 2.4
1146185296 M * daniel_hozac regardless of what you do, the chbind test would always fail.
1146185296 M * Bertl I'll check how it looks inside shortly :)
1146185380 M * Bertl okay, here is the conservative solution for stable:
1146185381 M * Bertl http://vserver.13thfloor.at/Experimental/delta-vxcapable-fix01a.diff
1146185383 M * doener daniel_hozac: only with legacy tools + recent kernel. Can we ignore that case?
1146185392 M * Bertl I tried to match the changes in devel as close as possible
1146185404 M * daniel_hozac doener: i thought that was the case we were working on.
1146185433 M * Bertl doener: would appreciate a deep look at the capability issues from your side, with focus on stable
1146185505 M * doener daniel_hozac: legacy config vs. legacy tools. legacy config + recent tools is what I assume to be used on 2.6 kernels. and then testme would not use the legacy tools
1146185520 M * Bertl indeed
1146185551 M * doener testme would fail with only legacy tools installed, indeed
1146185567 M * daniel_hozac ah right, it's a vxi capability.
1146185583 M * daniel_hozac err, flag.
1146185594 M * Bertl information :)
1146185607 M * Bertl and it's VCI :)
1146185619 M * doener ah, even that... so chbind alone would never cause any problems in that regard
1146185635 A * Bertl .o( why am I using such meaningfull names *G* )
1146185690 M * doener hm, i always wondered what vci means... guess now is the right time to ask :) so, what does vci mean?
1146185702 Q * weasel Read error: Operation timed out
1146185705 M * Bertl Vserver Configuration Information :)
1146185720 Q * Loki|muh Remote host closed the connection
1146185729 M * doener ah, configuration! always thought the C is "context" and then "virtual" made no sense
1146185745 M * doener and neither did "vserver"
1146185784 M * daniel_hozac doener: oh, and legacy config implies legacy tools. but testme doesn't use those.
1146185789 J * weasel weasel@asteria.debian.or.at
1146185794 M * Bertl doener: I had a longer name there but it didn't really look nice and I didn't consider this important in any way ...
1146185804 M * Bertl wb weasel!
1146185821 M * doener daniel_hozac: yeah, that's what i meant with "legacy tools vs. legacy config"
1146185822 M * Bertl btw, uploaded vcmd-0.03.tar.bz2
1146185836 Q * ag- jupiter.oftc.net neutron.oftc.net
1146185836 Q * nebuchadnezzar jupiter.oftc.net neutron.oftc.net
1146185836 Q * shedi jupiter.oftc.net neutron.oftc.net
1146185836 Q * VAndreas jupiter.oftc.net neutron.oftc.net
1146185836 Q * mountie jupiter.oftc.net neutron.oftc.net
1146185836 Q * jkl jupiter.oftc.net neutron.oftc.net
1146185842 M * Bertl does now allow to name the debug output commands for you, example:
1146185883 J * Loki|muh loki@satanix.de
1146185883 J * ag- ag@caladan.roxor.cx
1146185883 J * nebuchadnezzar ~nebu@zion.asgardr.info
1146185883 J * jkl eric@c-71-56-216-223.hsd1.co.comcast.net
1146185883 J * mountie ~mountie@CPEdeaddeaddead-CM000a739acaa4.cpe.net.cable.rogers.com
1146185883 J * VAndreas ~Hossa@212.110.98.7
1146185883 J * shedi ~siggi@inferno.lhi.is
1146185884 M * Bertl ./vcmd -dntC VCMD_52_4[0]
1146185885 M * Bertl VCMD_set_ccaps_v0 [v0]:
1146185925 M * weasel moo
1146185936 Q * ag- Remote host closed the connection
1146185953 J * ag- ag@caladan.roxor.cx
1146185982 Q * ray6 Ping timeout: 480 seconds
1146186009 M * daniel_hozac ah, cool.
1146186063 M * doener Bertl: capabilities issues being the cap-mask stuff?
1146186083 M * Bertl well, especially the delta-vxcapable-fix01a.diff
1146186090 M * Bertl to restrict ccaps to root
1146186265 M * doener ouch, did we actually give the ccaps to regular users as well?
1146186271 M * daniel_hozac yep.
1146186425 M * Bertl okay, uptime was not virtualized on legacy, so I leave that as is
1146186517 Q * gerrit Ping timeout: 480 seconds
1146186582 Q * brc Ping timeout: 480 seconds
1146186596 M * doener will take a look at it tomorrow (would appreciate a reminder ;) going to bed now
1146186603 M * doener good night folks!
1146186607 M * Bertl okay, night!
1146186614 Q * doener Quit: leaving
1146186642 Q * alamar_ Ping timeout: 480 seconds
1146187196 M * Bertl daniel_hozac: would it be hard to add the ipv4root info for non-legacy tools (as eyck mentioned)?
1146187202 M * daniel_hozac linux-2.6.16.11-vs2.1.1-rc17/include/linux/vs_context.h:{196,211} what is that supposed to do?
1146187219 M * daniel_hozac Bertl: it's already there, the tools are just called with --silent by default, IIRC.
1146187242 M * Bertl ah, cool, how to unsilence them?
1146187282 M * Bertl daniel_hozac: the enter/leave stuff?
1146187294 M * daniel_hozac vserver -v maybe? i'm not sure.
1146187299 M * daniel_hozac the xchg.
1146187310 M * daniel_hozac i get warnings for that, computed value is not used.
1146187352 M * Bertl interesting, could you upload them?
1146187363 M * Bertl the xchg basically exchanges two values in place
1146187387 M * daniel_hozac upload the warnings?
1146187406 M * Bertl yes, maybe with some context, or is it just that?
1146187414 M * daniel_hozac that's it.
1146187433 M * daniel_hozac include/linux/vs_context.h:211: warning: value computed is not used
1146187465 M * Bertl ah, try adding a (void) before the xchg
1146187520 M * Bertl what gcc is that 4.1.x?
1146187581 M * daniel_hozac gcc-4.1.0-3
1146187610 M * Bertl thought so, any other warnings?
1146187613 M * daniel_hozac it's just a warning, so it's no big deal.
1146187627 M * daniel_hozac none that i've noticed.
1146187643 M * daniel_hozac rather odd that the plm didn't pick up on that.
1146187653 M * Bertl they are on 4.0.x IIRC
1146187667 M * daniel_hozac i686/gcc4.1 Cross Compile
1146187690 M * Bertl okay ...
1146187746 M * Bertl http://vserver.13thfloor.at/Experimental/delta-xchg-clean01.diff
1146187749 M * Bertl does that fix it?
1146187770 M * daniel_hozac i would assume so.
1146187808 M * daniel_hozac what is line 95 supposed to do?
1146187812 J * brc bruce@20151185030.user.veloxzone.com.br
1146187861 M * Bertl it replaces the entry with NULL, retrieving the old info
1146187875 M * Bertl wb brc!
1146187881 M * daniel_hozac so basically short for x = y; y = NULL;?
1146187897 M * Bertl yeah, but it's atomic
1146187903 M * daniel_hozac ah.
1146187956 M * daniel_hozac yeah, no warnings now.
1146187962 M * Bertl great!
1146188009 M * Bertl I'm going to suggest the delta-vxcapable-fix01a.diff as fix for both stable and devel on the ML, do you agree that this should fix it?
1146188037 M * Bertl but I will put the new mbcap stuff in the next devel release
1146188065 M * Bertl (release candidate that is)
1146188071 M * daniel_hozac yeah, it should limit the ccaps to root.
1146188158 M * daniel_hozac i'll have the mbcap kernel ready for testing in a few minutes.
1146189314 M * Bertl vattribute was safe to use without --bcap as long as you do not use secure, right?
1146189335 M * Bertl or what was the bcap bug fixed some time ago?
1146189397 M * daniel_hozac hmm, i think it calls vc_set_ccaps either way.
1146189408 M * daniel_hozac -> you need to specify all the caps you want.
1146189424 M * Bertl what about vattribute --set --xid <xid> --bcap ~0 --ccap ~-1
1146189445 M * Bertl or is --bcap -1 better?
1146189461 M * daniel_hozac yeah, that'd work.
1146189526 J * Aiken_ ~james@tooax6-058.dialup.optusnet.com.au
1146189570 M * Bertl welcome Aiken_!
1146189679 M * Aiken_ good morning
1146189852 Q * Aiken Ping timeout: 480 seconds
1146189924 J * gerrit ~gerrit@c-67-160-146-170.hsd1.or.comcast.net
1146189978 J * matta ~matta@c-68-81-35-243.hsd1.pa.comcast.net
1146190233 M * Bertl wb gerrit! matta!
1146190544 M * daniel_hozac it seems to work fine (re: mbcap).
1146190559 M * Bertl excellent, what about bind?
1146190582 M * daniel_hozac my old test-case works flawlessly.
1146190597 M * Bertl okay, so we also solved the bind issue once for all?
1146190617 M * daniel_hozac it looks that way.
1146190621 M * daniel_hozac i'll install it and make sure.
1146190637 M * daniel_hozac yep, seems to start fine.
1146190677 M * Bertl great!
1146190699 M * daniel_hozac indeed.
1146191052 M * gerrit Rehi Bertl
1146191265 T * Bertl http://linux-vserver.org/ | latest stable 2.01, 1.2.10, 1.2.11-rc1, devel 2.1.0, exp 2.{0.2,1.1}-rc18 | util-vserver-0.30.210 | libvserver-1.0.2 & vserver-utils-1.0.3 | He who asks a question is a fool for a minute; he who doesn't ask is a fool for a lifetime -- share the gained knowledge on the wiki, and we'll forget about the minute ;)
1146191272 M * Bertl *-rc18
1146191327 M * daniel_hozac i've tested all the aspects i can think of. it behaves as expected.
1146191356 M * Bertl okay, I think we should especially look on things like netlink or nfs/quota
1146191382 M * Bertl will test some of the quota/nfs aspects shortly, maybe you could do some netlink tests?
1146191413 M * daniel_hozac such as? would ip do?
1146191425 M * Bertl yeah, something like this
1146191433 M * Bertl trying to change things and such
1146191485 M * Bertl I will check the code tomorrow once again if I can spot anything wrong there
1146191494 M * daniel_hozac humm.
1146191501 M * daniel_hozac ip addr add works inside a guest.
1146191520 M * Bertl ah, okay, so I'll check today :)
1146191865 M * daniel_hozac no vs_cvirt-clean01, vs_pid-clean01, or nopage-clean01?
1146191896 M * Bertl url?
1146191950 M * daniel_hozac http://daniel.hozac.com/vserver/delta-vs_cvirt-clean01.diff
1146191957 M * daniel_hozac http://daniel.hozac.com/vserver/delta-vs_pid-clean03.diff
1146191968 M * daniel_hozac http://daniel.hozac.com/vserver/delta-nopage-clean01.diff
1146192097 M * Bertl okay, the nopage will probably fall out when I break down the patches
1146192136 M * Bertl the other two are fine with me, one is against 2.0.2 the other 2.1.1?
1146192151 M * daniel_hozac yeah.
1146192193 M * Bertl do we know why the vs_cvirt was obsoleted in 2.0.2 in so many files? i.e. which include file pulls in that one?
1146192288 M * Bertl ah, this was the ptrace consolidation
1146192307 M * daniel_hozac yeah.
1146192439 M * Bertl okay, but proc is wrong
1146192444 M * Bertl kernel/vserver/proc.c: In function `proc_xid_cvirt':
1146192445 M * Bertl kernel/vserver/proc.c:153: warning: implicit declaration of function `vx_update_load'
1146192486 M * daniel_hozac hmm, did it remove that?
1146192489 M * Bertl did you test those somehow?
1146192492 M * daniel_hozac it wasn't supposed to...
1146192501 Q * Aiken_ Quit: Leaving
1146192513 M * daniel_hozac well, this batch was autogenerated.
1146192526 M * Bertl and not compile tested, I suppose?
1146192530 M * daniel_hozac i thought i had fixed the script to remove all the bogus results, but i apparently missed that one.
1146192552 M * daniel_hozac not yet, no.
1146192580 M * Bertl okay, I put it in a temp version and let it run through PLM
1146192604 J * Aiken ~james@tooax6-058.dialup.optusnet.com.au
1146192622 M * daniel_hozac ah, i had missed part of an if.
1146192642 M * Bertl so, new upcoming patch set?
1146192682 M * daniel_hozac nah, it'd just remove the kernel/vserver/proc.c hunk in the vs_cvirt patch.
1146192694 M * Bertl okay, and for the other one?
1146192732 M * daniel_hozac those should all be ok.
1146192737 M * Bertl okay
1146192749 M * daniel_hozac previously i had manually removed all the files that needed linux/vserver/cvirt.h.
1146192772 M * daniel_hozac but that logic is only needed on 2.0, as 2.1 has the pid stuff separated.
1146192827 M * Bertl ah, upcoming fix for the netlink stuff :)
1146192839 M * daniel_hozac great! :)
1146192872 M * Bertl I did fix dummy caps, but not commoncap :)
1146193073 M * daniel_hozac well, i think i'm gonna get some sleep now. i'll continue testing tomorrow.
1146193080 M * daniel_hozac good night!
1146193087 M * Bertl okay, good night!
1146194058 J * alamar ~alamar@sol.diddens.de
1146194071 M * Bertl welcome alamar!
1146194232 M * Bertl okay, off to bed now too, have a good once everyone!
1146194242 N * Bertl Bertl_zZ
1146197642 J * _coocoon_ ~coocoon@p54A0547C.dip.t-dialin.net
1146197657 M * _coocoon_ morning
1146198605 M * eyck Bertl: pong
1146204995 Q * id23 Ping timeout: 480 seconds
1146205001 J * knotty ~void@loin.ailleurs.org
1146205537 J * id23 ~id@p54A050F8.dip0.t-ipconnect.de
1146205872 J * kilgur user@p50811820.dip0.t-ipconnect.de
1146205878 M * kilgur hi
1146207418 M * harry there we go... we've got kernel 2.6.16.11, grsec 2.1.9 and vserver 2.0.2-rc18
1146207421 M * harry lets merge!
1146207883 J * chelli ~thomas@PC1.ideolabs.TGZ-Ilmenau.de
1146207905 N * otaku42_away otaku42
1146208067 J * Aiken_ ~james@tooax8-074.dialup.optusnet.com.au
1146208382 Q * Aiken Ping timeout: 480 seconds
1146208495 J * bricecube ~bricecube@va-71-51-0-34.dhcp.sprint-hsd.net
1146208663 N * BobR_oO BobR
1146208680 J * davut ~davut@213.74.180.7
1146208709 M * davut hi
1146208716 M * BobR hi
1146208741 M * davut I thought #vserver was on freenode :)
1146208867 M * davut I am a gentoo user and I wanna use vserve + grsec
1146208900 M * davut in the changelog of vserver ebuild it says there is grec patch but I couldnt find it
1146208921 M * davut aynyway is there any gentoo vserver user using grsec?
1146209266 J * ||Cobra|| ~cob@pc-csa01.science.uva.nl
1146209726 J * pagano ~pagano@lappagano.cnaf.infn.it
1146210317 M * Hollow davut: there are no grsec patches in vserver-sources since 1.9* or so..
1146210336 M * Hollow it's just too much a pita to merge grsec + vserver
1146210349 M * davut hmmm
1146210369 M * davut Hollow: how can I use grsec+vserver together
1146210376 M * Hollow harry: do you know any recent grsec+vserver patches?
1146210411 M * Hollow davut: i don't know, never used grsec
1146210416 M * Hollow but if you find a patch
1146210423 M * Hollow there maybe patches for some older versions
1146210630 M * harry Hollow: i'm working on it
1146210633 M * harry 08:56 < harry> there we go... we've got kernel 2.6.16.11, grsec 2.1.9 and vserver 2.0.2-rc18
1146210638 M * harry will be there later today...
1146210661 M * Hollow davut: there you go
1146210674 M * harry davut: if you want: http://harry.ulyssis.org/vserver
1146210694 M * harry the latest is 2.6.14.7 (because there was no grsec for 2.6.15 and 2.6.16
1146211347 M * davut hm ok
1146211355 A * harry has 1 big q!
1146211371 M * harry + static int chown_common(struct dentry *dentry, struct vfsmount *mnt,
1146211371 M * harry +   uid_t user, gid_t group)
1146211391 M * harry why can't we put the struct vfsmount *mnt as LAST argument?
1146211420 M * harry like they do in grsec
1146211433 M * harry it would make the patch merge easier
1146211448 M * harry and it would also make more sense to add own parameters at the end
1146212854 Q * cryo Ping timeout: 480 seconds
1146213103 J * cryo ~say@psoft.user.matrix.farlep.net
1146213613 Q * shedi Quit: Leaving
1146214238 N * BobR BobR_oO
1146214449 J * pagano_ ~pagano@lappagano.cnaf.infn.it
1146214480 Q * pagano Read error: Connection reset by peer
1146216690 Q * mkhl Ping timeout: 480 seconds
1146218213 M * otaku42 moin
1146218226 M * otaku42 question: am i right that klogd is only needed on the host, not on the guests?
1146218414 M * Hollow otaku42: right
1146218521 M * otaku42 Hollow: and is there an easy way to avoid the need for running syslog in every guest, when i have syslogd on the host and want that to gather the logs from the guests in a central place?
1146218769 M * Hollow well, you can log everything to one syslog if you use tcp/udp as destination in the guests, and tcp/udp as source on the host
1146218783 M * Hollow but you still need to run a syslog in the guest
1146218797 M * Hollow and i'm not sure if syslog can log to tcp.. i'm using syslog-ng everywhere
1146219123 M * otaku42 Hollow: not sure either, but udp would serve as well (and that works with syslog(non-ng))
1146219153 M * Hollow yup
1146219164 M * Hollow as long as some network connection is there :P
1146219179 M * otaku42 Hollow: since i also have a private network up that all guests and the host are connected to, that should be fine
1146219207 M * otaku42 Hollow: if dummy0 is not available anymore, i guess i will have other problems than not being able to get any logs from the guests :)
1146219214 M * Hollow heh
1146219226 M * DEac- i want to start an other init in a vserver. then i want to see what it prints. how i can see the output?
1146219292 M * Hollow you can't see the output of init
1146219299 M * DEac- no way?
1146219305 M * Hollow exactly
1146219317 M * DEac- bad
1146219318 M * Hollow (except your init scripts log to a file)
1146219324 M * Hollow but the plain output of init won't show up
1146219330 M * Hollow as this goes to /dev/console
1146219339 M * Hollow which is not available, so it goes to /dev/null
1146219361 M * DEac- ah, i can pipe it
1146219383 M * otaku42 Hollow: i vaguely remember that there was an option to redirect /dev/console of a guest to a "real" console on the host. not sure if i remember correctly, and no idea where i've read that, though
1146219428 M * Hollow well, one can use the vlogin mechanism to create a new pseudo terminal and let init use this pts, but you'll never know when boot up finished, because init runs forever, so the "vserver ... start" process will never finish
1146219457 M * DEac- Hollow: that's ok
1146219458 M * Hollow and, be honest... who needs to view inits output?
1146219462 M * Hollow you have logging for that
1146219501 M * DEac- Hollow: i need it. i'm a developer of initng and it's the init-process
1146219540 M * Hollow well, then use the vlogin stuff and patch initng to use a pts instead of /dev/console
1146219581 M * Hollow i have not tested it, but in theory it should work
1146219620 M * DEac- i think, i try it with a pipe
1146219674 M * DEac- oh, it use it for in and output, bad
1146219834 M * Hollow hm?
1146219902 M * DEac- in and output with one pipe isn't possible
1146219911 M * DEac- it reads, what it writes
1146219934 M * Hollow vlogin ;)
1146219948 M * Hollow it is important to allocate a new pts _inside_ the context
1146219954 M * Hollow else you won't be able to open it
1146220070 M * DEac- Hollow: then i must patch initng, and i don't want. i want to use it nativ, unpatched
1146220181 M * Hollow well, it won't work, as it forks to background, so you have no control over it on the host anymore, i.e. you won't get the output
1146220209 M * Hollow the vserver ... start process finishes and if you allocate the pts or not, you won't see it
1146220247 M * Hollow and you also don't want to not fork it
1146220253 M * DEac- console must be an other dev then console. a dev, which vserver provides
1146220279 M * Hollow there is none
1146220326 J * Lostfrog ~Void@203.33.121.211
1146220387 Q * davut Quit: davut
1146220388 M * Lostfrog does vserver like linuxvirtualserver.org??? 
1146220393 M * Lostfrog is
1146220446 M * Hollow no, completely different thing
1146220480 M * Lostfrog hmmm is it for shared hosting?
1146220483 M * Hollow linuxvirtualserver.org combines many machines to one, linux-vserver splits one machine into many :)
1146220513 M * Lostfrog cheers
1146220518 M * Lostfrog good thing
1146220568 M * Hollow afk for a bit
1146220809 J * cehteh foobar@217.140.77.75
1146221935 Q * TheSeer Quit: Client exiting
1146222000 Q * softi42 Ping timeout: 480 seconds
1146222503 Q * id23 Quit: Leaving
1146222556 J * softi42 ~softi@p549D798C.dip.t-dialin.net
1146222882 Q * pagano_ Ping timeout: 481 seconds
1146223326 Q * _coocoon_ Quit: KVIrc 3.2.0 'Realia'
1146225067 Q * Aiken_ Ping timeout: 480 seconds
1146225078 J * pagano ~pagano@lappagano.cnaf.infn.it
1146225743 J * _coocoon_ ~coocoon@p54A0547C.dip.t-dialin.net
1146226244 P * matti 8-X
1146226258 Q * tokkee Remote host closed the connection
1146227061 J * aphetadus ~joaopaulo@149121.cps.virtua.com.br
1146227096 J * tokkee tokkee@213.95.10.25
1146227564 Q * kilgur Ping timeout: 480 seconds
1146228293 Q * matta Ping timeout: 480 seconds
1146228327 Q * DEac- Ping timeout: 480 seconds
1146228996 J * DEac- ~deac@xdsl-195-14-216-254.netcologne.de
1146229161 J * kilgur user@p50811820.dip0.t-ipconnect.de
1146229567 J * matta ~matta@c-68-32-239-173.hsd1.pa.comcast.net
1146230524 Q * Loki|muh Ping timeout: 480 seconds
1146230721 J * Loki|muh loki@satanix.de
1146230734 P * aphetadus 
1146230970 N * Bertl_zZ Bertl
1146230978 M * Bertl morning folks!
1146230990 M * _coocoon_ morning bertl
1146231028 M * Bertl hey _coocoon_!
1146231115 J * dearaujo ~dan@pixpat.austin.ibm.com
1146231210 M * Bertl welcome dearaujo!
1146231216 M * waldi hi Bertl 
1146231219 M * dearaujo good morning
1146231229 M * dearaujo /evening
1146231389 M * Bertl okay, off for now ... back later ...
1146231398 N * Bertl Bertl_oO
1146233536 M * harry wiiiiiii
1146234836 J * matt1 ~matta@c-68-32-239-173.hsd1.pa.comcast.net
1146234995 Q * pagano Ping timeout: 480 seconds
1146235053 J * Viper0482 ~Viper0482@p54974F04.dip.t-dialin.net
1146235264 Q * chelli Quit: Verlassend
1146235269 Q * matta Ping timeout: 480 seconds
1146235791 N * matt1 matta
1146236715 P * kilgur 
1146236827 Q * insomniac Remote host closed the connection
1146237351 N * Bertl_oO Bertl
1146237364 M * Bertl short visit ~30min
1146237441 N * otaku42 otaku42_away
1146237721 Q * restill Ping timeout: 480 seconds
1146238145 M * daniel_hozac Bertl: 2.1.1-rc18, ./arch/um/sys-x86_64/syscalls.c is missing linux/vs_cvirt.h
1146238194 M * Bertl okay, patch?
1146238233 J * meandtheshell ~markus@85-125-230-9.dynamic.xdsl-line.inode.at
1146238251 M * W0nka -rc18?
1146238252 M * W0nka wtf?
1146238261 M * W0nka http://linux-vserver.org/ still says -rc16
1146238287 M * daniel_hozac http://daniel.hozac.com/vserver/delta-uml-fix01.diff
1146238296 M * Bertl W0nka: for up-to-date info, see topic
1146238315 M * W0nka yeah, saw it there...
1146238321 M * W0nka but why is it not on the website?
1146238322 M * daniel_hozac or http://vserver.13thfloor.at/Experimental/
1146238328 M * daniel_hozac because nobody has updated it yet.
1146238331 M * daniel_hozac feel free ;)
1146238362 M * Bertl daniel_hozac: btw, uploaded 5106/7_cross.diff
1146238372 M * daniel_hozac ah, done already?
1146238373 M * W0nka *dl'ing http://vserver.13thfloor.at/Experimental/patch-2.6.16.11-vs2.0.2-rc18.diff*
1146238374 M * daniel_hozac that was quick.
1146238384 Q * knotty Quit: Parti Ailleurs
1146238399 M * Bertl yep, got the results per email ...
1146238409 M * daniel_hozac cool.
1146238445 J * mkhl ~mkhl@200-153-181-123.dsl.telesp.net.br
1146238466 M * daniel_hozac do we care about the Using plain integer as NULL pointer warnings?
1146238498 M * Bertl well, we might fix that sooner or later
1146238684 M * daniel_hozac the kernel/sys.c:519 warnings look legit. vs_reboot probably should declare the second argument as __user too.
1146238729 M * daniel_hozac (not that it's used or anything)
1146238803 M * daniel_hozac (for 5106, 2.0.2-rc18)
1146238833 M * Bertl doing a full report now, just to make sure we did not miss anything
1146238848 J * insomniac ~insomniac@slackware.it
1146238874 M * Bertl welcome insomniac!
1146238935 M * insomniac hi Bertl :)
1146239021 M * _coocoon_ need the vserver ... build switch that the failled installation will not be removed
1146239032 M * daniel_hozac --keep
1146239037 M * daniel_hozac as vserver - build --help says.
1146239077 M * _coocoon_ ok cool thanx
1146239146 M * waldi daniel_hozac: which warning?
1146239163 M * daniel_hozac re: kernel/sys.c?
1146239168 M * waldi daniel_hozac: it is valid C to use 0 in pointer context
1146239179 M * daniel_hozac waldi: yes, i know. that's why i asked if we cared at all :)
1146239201 M * daniel_hozac http://vserver.13thfloor.at/Experimental/5106_cross.diff
1146239209 M * daniel_hozac (5107 for devel)
1146239210 M * waldi (and in fact, glibc defines NULL as 0 at least for c++ code)
1146239229 M * daniel_hozac IIRC it's ((void*)0) for C.
1146239262 M * Bertl note: moved all the cross reports to PLM_CHECK
1146239274 M * daniel_hozac ok.
1146239285 M * Bertl there is also a *cross_full* now for 5106/7
1146239304 M * waldi daniel_hozac: for C it is a void pointer, correct
1146239328 M * Bertl moving out now .. will be back later this evening ...
1146239335 J * WorkRoey ~katz@h-69-3-4-130.mclnva23.covad.net
1146239362 M * harry davut, Hollow : the vserver + grsec patch are online
1146239367 M * WorkRoey W0nka:  oh there you are
1146239367 M * WorkRoey hi!
1146239369 M * WorkRoey harry:  yo
1146239372 N * Bertl Bertl_oO
1146239373 M * WorkRoey Bertl:  hey Bertl!
1146239380 N * W0nka Wonka
1146239389 M * Wonka .oO( where did that 0 come from?!? )
1146239392 M * harry btw, Bertl : why the allignment of the TASK_SIZE in ELF_ET_DYN_BASE?
1146239395 M * Wonka hi roey!
1146239491 M * WorkRoey Wonka:  hey hey
1146239515 M * WorkRoey Wonka:  ok, so I've found another openvpn/vserver wannitnow guy
1146239526 M * Wonka not really
1146239536 M * Wonka as i said, openvpn is another construction site :)
1146239561 M * Wonka openvpn is on the router box, not on the vserver host...
1146239566 M * WorkRoey does openvpn know that they need to seriously improve on some things before normal shops (like mine) can use it??
1146239574 P * dearaujo 
1146239622 M * harry i've been talking to pax people
1146239650 M * WorkRoey For example, this business of requiring admin privs on windows clients... I don't want to give employees that kind of leeway on a company laptop.
1146239660 M * WorkRoey harry:  what did the pax people say?
1146239675 M * harry there is an integer overflow possible in vserver
1146239688 M * WorkRoey as of which version??
1146239695 M * harry all of them off course
1146239704 M * harry if you get TASK_UNMAPPED_BASE > 2GB
1146239710 M * WorkRoey oh.. ok.
1146239719 M * harry you define ELF_ET_DYN_BASE as ((TASK_UNMAPPED_BASE) * 2)
1146239722 M * WorkRoey well thanks for letting me know; now I'll cringe and wait for an updated version :)
1146239725 M * harry which results in an integer overflow
1146239745 M * Wonka WorkRoey: afair, in 2.1, they don't need admin privs any more
1146239778 Q * ||Cobra|| Remote host closed the connection
1146239780 M * WorkRoey Wonka:  of openvpn
1146239780 M * WorkRoey ?
1146239784 M * Wonka WorkRoey: yes
1146239793 M * WorkRoey Wonka:  how so?
1146239808 M * Wonka WorkRoey: dunno. just read it.
1146239813 M * Wonka WorkRoey: but let's keep openvpn stuff over there...
1146239817 M * WorkRoey oh so this is a brand-new thing??
1146239818 M * WorkRoey ok
1146240561 M * daniel_hozac harry: how would you get that?
1146240585 M * harry not sure yet ;)
1146240603 M * harry > so if i'm correct here, it's exactly the same as just removing the
1146240603 M * harry > segnexec stuff and just #define ELF_ET_DYN_BASE        
1146240603 M * harry > ((TASK_UNMAPPED_BASE) * 2)
1146240603 M * harry yes, that should work, although i personally don't agree with the tying
1146240603 M * harry of ELF_ET_DYN_BASE to TASK_UNMAPPED_BASE because the two are conceptually
1146240605 M * harry quite independent. not to mention the potential for a nice integer
1146240608 M * harry overflow if someone sets TASK_UNMAPPED_BASE above 2GB (historically this
1146240611 M * harry define had such a problem in fact, it used to be defined as TASK_SIZE*2/3
1146240613 M * harry and not TASK_SIZE/3*2, resulting in an unintended address).
1146240637 M * harry so normally... it won't be a problem
1146240970 M * daniel_hozac unless my math coprocessor has gone nuts, i don't see how you could ever get a PAGE_OFFSET above 6 GiB...
1146241000 M * daniel_hozac (which would be required to get a 2 Gi TASK_UNMAPPED_BASE)
1146241024 M * harry true again...
1146241046 A * harry no expert here... but what's the advantage in alligning TASK_SIZE
1146241068 M * daniel_hozac you want to have things start at a page boundary.
1146241105 M * harry hmm..
1146241130 M * harry then why don't they do that in a vanilla kernel?
1146241156 M * daniel_hozac because vanilla has a completely different method of calculating the TASK_UNMAPPED_BASE?
1146241188 M * daniel_hozac hmm, no, it too has PAGE_ALIGN(TASK_SIZE / 3).
1146241217 M * harry that's what i thought ;)
1146241225 M * harry the only difference is the allignment
1146241235 M * daniel_hozac hmm?
1146241304 M * daniel_hozac what alignment? what are you talking about?
1146241327 M * harry the page_align of task_size
1146241386 M * harry #define TASK_UNMAPPED_BASE  (PAGE_ALIGN(TASK_SIZE / 3))
1146241419 M * harry -#define ELF_ET_DYN_BASE         (TASK_SIZE / 3 * 2)
1146241419 M * harry +#define ELF_ET_DYN_BASE        ((TASK_UNMAPPED_BASE) * 2)
1146241425 M * harry the last is the vserver patch
1146241471 M * daniel_hozac ah, you mean that.
1146241604 M * daniel_hozac i guess that should be in vanilla.
1146241691 M * harry that's why i think it's rather strange ;)
1146241735 P * Lostfrog 
1146241764 M * daniel_hozac hmm, well, it seems all uses of it are wrapped in ELF_PAGESTART()s.
1146241806 M * harry well... the problem now is... i don't know what to do best...
1146241817 M * harry page_align task_size or not
1146242053 M * daniel_hozac doesn't really matter, as the ELF code aligns it when it's used anyhow
1146242060 M * daniel_hozac (in that one location...)
1146242079 M * harry so i'll just use TASK_SIZE/3*2
1146242105 M * harry don't do the page_align
1146242348 M * _coocoon_ daniel_hozac: do u know this error message during installing fc5 on x86_64 architecture -->Transaction Check Error:   package glibc-2.4-4 is already installed
1146242377 M * _coocoon_ daniel_hozac: vserver building
1146242440 M * daniel_hozac during FC5 install? or when building an FC5 guest?
1146242467 M * _coocoon_ daniel_hozac:when building a guest sorry
1146242473 M * daniel_hozac on what kind of host?
1146242508 M * _coocoon_ daniel_hozac: distribution or what do u mena
1146242516 M * daniel_hozac yes.
1146242525 M * _coocoon_ daniel_hozac: suse 10.0 amd64
1146242547 M * _coocoon_ 2.6.16.8-vs2.1.1-rc17-default
1146242574 M * daniel_hozac what yum are you using?
1146242577 M * _coocoon_ util-vserver: 0.30.210
1146242611 M * _coocoon_ 2.4.0
1146242621 M * daniel_hozac with the patch, i guess?
1146242624 Q * Viper0482 Ping timeout: 480 seconds
1146242630 M * _coocoon_ no without
1146242652 M * _coocoon_ ok so i need a patch for fc5 but all other dists fc3/4 and centos4.3 works fine
1146242657 Q * brc Quit: [BX] 2000: year of the BitchX
1146242696 J * brc_ bruce@20151185030.user.veloxzone.com.br
1146242706 M * daniel_hozac what does your configuration look like? have you enabled multilib?
1146242732 M * _coocoon_ daniel_hozac: hm where can i activate it
1146242756 M * daniel_hozac i'm not sure. i guess you're trying to install glibc i386 and x86_64 at the same time, right?
1146242786 M * _coocoon_ daniel_hozac this is the problem right i won't try but yum does it
1146242805 M * _coocoon_ in yum config file
1146242808 M * _coocoon_ ?
1146242931 M * daniel_hozac do you have exactarch=1 in your yum.conf?
1146242940 M * daniel_hozac (the one the utils will use for building the guest)
1146243112 J * Viper0482 ~Viper0482@p54977E50.dip.t-dialin.net
1146243292 Q * _coocoon_ Ping timeout: 480 seconds
1146243863 J * _coocoon_ ~coocoon@p54A0635F.dip.t-dialin.net
1146244836 J * doener ~doener@i5387DE42.versanet.de
1146245301 Q * Hollow Quit: Konversation terminated!
1146245464 J * Hollow ~hollow@home.xnull.de
1146246107 Q * ntrs Ping timeout: 480 seconds
1146246472 J * shedi ~siggi@inferno.lhi.is
1146246574 Q * bricecube Quit: Leaving
1146247097 J * ntrs ~ntrs@dsl-203-81-rt2-bras.wnvl.centurytel.net
1146247445 Q * Snow-Man Ping timeout: 480 seconds
1146247621 J * Snow-Man ~sfrost@kenobi.snowman.net
1146250437 Q * shedi Read error: Connection reset by peer
1146251222 J * shedi ~siggi@inferno.lhi.is
1146251257 Q * ntrs Read error: Connection reset by peer
1146251273 J * ntrs ~ntrs@dsl-203-81-rt2-bras.wnvl.centurytel.net
1146252523 Q * meandtheshell Quit: bye bye ... 
1146253695 Q * mnemoc Ping timeout: 480 seconds
1146254329 J * mnemoc ~amery@216.241.24.10
1146254465 J * ntrs_ ~ntrs@68-188-51-87.dhcp.stls.mo.charter.com
1146254551 Q * shedi Quit: Leaving
1146254728 J * bonbons ~bonbons@83.222.39.134
1146254911 Q * ntrs Ping timeout: 480 seconds
1146255197 J * kir_home ~kir@ffm-nat.sw-soft.de
1146255391 P * matta 
1146255971 Q * kir_home Quit: Ухожу я от вас
1146257864 M * WorkRoey ciao all
1146257867 P * WorkRoey Leaving
1146258316 T * * http://linux-vserver.org/ | latest stable 2.01, 1.2.10, 1.2.11-rc1, devel 2.1.0, exp 2.{0.2,1.1}-rc18 | util-vserver-0.30.210 | libvserver-1.0.2 & vserver-utils-1.0.3 | He who asks a question is a fool for a minute; he who doesn't ask is a fool for a lifetime -- share the gained knowledge on the wiki, and we'll forget about the minute ;)
1146258316 T * Bertl -
1146258378 Q * Bertl_oO Remote host closed the connection
1146258379 J * Bertl_oO herbert@212.16.62.52
1146258385 Q * BobR_oO Remote host closed the connection
1146258400 J * BobR_oO odie@212.16.62.52
1146258401 Q * morrigan Remote host closed the connection
1146258402 J * morrigan morrigan@212.16.62.52
1146258498 Q * Viper0482 Read error: Connection reset by peer
1146259076 P * _coocoon_ So Long, and Thanks for All the Fish!
1146259566 J * Pater_John ~Patrick09@194.112.144.69
1146259588 M * Pater_John hello @ all
1146259604 M * Pater_John i have one problem, can someone help me please?
1146259614 M * Pater_John Bringing up loopback interface: Device lo does not seem to be present, delaying initialization.
1146259614 M * Pater_John [FAILED]
1146259634 M * Pater_John what issue can this cause?
1146259762 J * _coocoon_ ~coocoon@p54A0635F.dip.t-dialin.net
1146260049 M * doener hardware related stuff is not allowed in vservers. and as networking happens on the host, you don't have to worry about that failure, you can happily remove that script from your runlevels
1146260199 M * Pater_John ah, ok
1146260202 Q * bonbons Quit: Leaving
1146262952 Q * _coocoon_ Ping timeout: 480 seconds
1146263535 J * _coocoon_ ~coocoon@p54A0595B.dip.t-dialin.net
1146263639 J * shedi ~siggi@inferno.lhi.is
1146264592 Q * Pater_John Ping timeout: 480 seconds
1146265097 N * Bertl_oO Bertl
1146265101 M * Bertl evening folks!
1146265119 M * daniel_hozac evening.
1146265155 M * doener hi Bertl 
1146265158 M * daniel_hozac http://daniel.hozac.com/vserver/delta-plm-fix02.diff
1146265357 M * Bertl interesting ...
1146265427 M * daniel_hozac should fix the "important" sparse compilation warnings/errors for 2.0.2.
1146265540 M * daniel_hozac i just fixed the "Should it be static?" for the do_* functions (as do_* has a very internal feel to it).
1146265691 M * Bertl okay, yes, I agree with the do_ statics
1146265704 M * Bertl they are not intended for global reference
1146266093 M * Bertl ah, you changed the ops to c99, good!
1146266136 M * daniel_hozac ah right, i wanted to verify that. i'm not sure exactly what the C99 way is :)
1146266185 M * Bertl looks fine to me, but haven't bother to check/change
1146266199 M * Bertl thanks a lot! 
1146266227 M * daniel_hozac i test compiled it with make allyesconfig, so it compiles on i686 at least.
1146266278 M * Bertl that should be sufficient, with the 4.1.x gcc?
1146266300 M * daniel_hozac yep, it's all i've got.
1146266321 M * Bertl is 4.1 officially released, btw?
1146266348 M * daniel_hozac Current release series:  GCC 4.1.0 (released 2006-02-28)
1146266364 M * Bertl ah, good, will try to make a cross compile set for that
1146266514 M * Bertl btw, I'm trying (for some time now, but only very sporadically) to build a bash with dietlibc (i.e. recent bash with all features) do you (aor anybody here) have any experience with that?
1146266662 M * daniel_hozac hmm, no, sorry.
1146266671 M * Bertl np, just asking ...
1146266734 M * Bertl ah, does the netlink fix (i.e. 2.1.1-rc18) work for you?
1146266765 M * daniel_hozac yeah.
1146266795 M * daniel_hozac i get EPERM when i try to add the address now :)
1146267557 M * daniel_hozac http://daniel.hozac.com/vserver/delta-plm-fix02d.diff for devel. still test-compiling it.
1146267634 M * daniel_hozac is the cast in mm/filemap.c needed at all? it's void __user *.
1146268293 M * doener OT - I'm lost in the new netfilter config options... :( what do I need for nat support?
1146268439 M * doener ah, nvm... didn't check back into the second submenu after fixing the first one... now the options are all there...