1226966569 Q * dowdle Remote host closed the connection
1226967076 Q * pisco Remote host closed the connection
1226967134 J * pisco ~pisco@86.59.118.153
1226967457 Q * Mojo1978 Ping timeout: 480 seconds
1226967558 J * Mojo1978 ~Mojo1978@ip-88-152-50-100.unitymediagroup.de
1226968050 Q * hparker Quit: Quit
1226968181 Q * pisco Ping timeout: 480 seconds
1226968259 J * pisco ~pisco@86.59.118.153
1226968570 J * hparker ~hparker@2001:470:1f0f:32c:212:f0ff:fe0f:6f86
1226968765 Q * geb Remote host closed the connection
1226970393 Q * bonbons Remote host closed the connection
1226970439 J * xdr ~xdr@118-173-96-87.cust.blixtvik.se
1226970546 Q * xdr_ Ping timeout: 480 seconds
1226973993 J * ntrs_ ~ntrs@77.29.21.43
1226974477 Q * ntrs_ Ping timeout: 480 seconds
1226975358 Q * nenolod Quit: Leaving
1226975582 J * nenolod ~nenolod@ip70-189-74-62.ok.ok.cox.net
1226978219 J * balbir ~balbir@32.97.110.53
1226978455 Q * doener Quit: leaving
1226979106 J * doener ~doener@i577AF9EF.versanet.de
1226979834 Q * Mojo1978 Read error: Connection reset by peer
1226980158 Q * ensc Ping timeout: 480 seconds
1226984221 J * ensc ~irc-ensc@p54B4E1DE.dip.t-dialin.net
1226984428 Q * hparker Quit: Read error: 104 (Peer reset by connection)
1226984731 N * quinq qzqy
1226985308 J * hparker ~hparker@2001:470:1f0f:32c:212:f0ff:fe0f:6f86
1226990596 J * sharkjaw ~gab@149-67-194.231210.adsl.tele2.no
1226991944 N * Bertl_zZ Bertl_oO
1226992680 J * mtg ~mtg@vollkornmail.dbk-nb.de
1226993685 J * davidkarban ~david@193.85.217.71
1226994246 Q * larsivi Ping timeout: 480 seconds
1226995581 Q * grobie` Ping timeout: 480 seconds
1226997226 J * larsivi ~larsivi@85.221.53.194
1226997976 J * arapaho ~arapaho@213.223.114.206
1227000369 Q * bibabu Quit: Coyote finally caught me
1227000373 J * bibabu bibabu@mikene.org
1227000526 J * grobie ~grobie@valgrind.schnuckelig.eu
1227001118 Q * hparker Quit: Read error: 104 (Peer reset by connection)
1227001446 N * morrigan_oO morrigan
1227001741 J * Hurga ~foest@h-213.61.155.114.host.de.colt.net
1227001791 N * arapaho Guest3848
1227001806 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d
1227001846 Q * Guest3848 Ping timeout: 480 seconds
1227001866 J * arapaho ~arapaho@213.223.114.206
1227001942 M * Hurga cool, just entering the channel makes me realize my mistake
1227002061 M * mnemoc that's Bertl-effect, one gets enlightened just by joining :D
1227002085 M * Hurga could be :)
1227002690 J * chI6iT41 ~chigital@services.mivitec.net
1227002808 M * pmenier Hello
1227002832 M * pmenier Problem with last patch ?  http://paste.linux-vserver.org/12606
1227003151 Q * pmenier Read error: Connection reset by peer
1227003223 J * pmenier ~pme@LNeuilly-152-22-72-5.w193-251.abo.wanadoo.fr
1227003374 J * _nono_ ~gomes@libation.ircam.fr
1227003517 M * bonbons pmenier: seems so, I'm seeing same behavior here ...
1227003563 M * bonbons seems any non-zero context as no /proc access anymore
1227003579 M * pmenier ah ok... i just reboot on 2.6.27.5-vs2.3.0.35.9 to be sure :)
1227003614 M * arekm bonbons: will test ipv6 patch later
1227003737 M * bonbons arekm: for what I could test here with the patch bind(<neighbor's addr>) now fails [did no attempt without it]
1227006306 M * fb pmenier,bonbons: upgrade your util-vserver
1227006597 M * pmenier what version ? i use 0.30.216-pre2794
1227006706 M * pmenier will try pre2824
1227007131 M * pmenier same error with 0.30.216-pre2824
1227008662 Q * eyck Ping timeout: 480 seconds
1227009670 J * dna ~dna@52-200-103-86.dynamic.dsl.tng.de
1227009744 J * eyck 08fkGSw8@nat03.nowanet.pl
1227010752 J * Mojo1978 ~Mojo1978@ip-88-152-50-100.unitymediagroup.de
1227011826 Q * eyck Ping timeout: 480 seconds
1227012307 J * eyck xC4172wW@nat05.nowanet.pl
1227013347 J * derjohn_mob ~aj@139.12.1.252
1227013409 J * wibble wibble@vortex.ukshells.co.uk
1227014839 J * kir ~kir@swsoft-msk-nat.sw.ru
1227014871 Q * sharkjaw Remote host closed the connection
1227014889 M * arekm bonbons: bind is disallowed now and correct source is used if I do ping6 some_remote_v6
1227014923 M * arekm bonbons: netlink still lies but this is the same af for ipv4
1227014997 M * bonbons arekm: netlink? what's the test-case?
1227015029 M * arekm bash-3.2# ip r g 1.1.1.1
1227015029 M * arekm 1.1.1.1 via 192.168.0.254 dev eth0  src 192.168.0.250 cache  mtu 1500 advmss 1460 hoplimit 64
1227015033 M * arekm bash-3.2# ip a |grep 250
1227015035 M * arekm bash-3.2#    
1227015036 M * arekm see "src 192.168.0.250"
1227015044 M * arekm (that's from inside of guest)
1227015113 N * qzqy quinq
1227015310 M * bonbons arekm: cool, on my old kernel the route case if correct for ipv6 but not for ipv4
1227015338 M * bonbons that is ipv6 suggests guest's source address, but ipv4 suggests host's main address
1227015378 M * arekm bonbons: is the "src xx" address on host the same as shown in guest by any luck?
1227015391 M * arekm bash-3.2# ip a |grep :200
1227015391 M * arekm bash-3.2# ip r g 2a01:390:1:0:a800:ffff:fede:ad05
1227015391 M * arekm 2a01:390:1:0:a800:ffff:fede:ad05 from :: via fe80::211:22ff:fe33:4455 dev eth0  proto kernel  src 2002:594c:1b49:1:211:d8ff:feb3:200  metric 1024  expires 27sec mtu 1280 advmss 1220 hoplimit 64
1227015445 M * bonbons yes, matches host/guest in all cases ... will check in a second guest
1227015575 M * bonbons everywhere the same, so the first on seems to have been winning... any later attempt comes from some cache
1227015606 M * arekm then try different ip in each guest :)
1227015637 Q * opuk Quit: leaving
1227015652 M * arekm here wrong src is displayed at netlink but of course correct one when actually trying to connect
1227015708 M * bonbons checking which address is used, it's the first one listed by ip addr list for both ipv4 and ipv6
1227015761 M * bonbons I would assume nxi is not passed for the netlink case... (I hope the source selection code is not duplicated for this!)
1227015771 A * bonbons out to lunch
1227016648 M * daniel_hozac arekm: we don't modify routes.
1227016899 M * arekm so "broken by design"
1227016963 M * arekm reboothing then to new fixed kernel
1227019858 Q * larsivi Remote host closed the connection
1227020653 Q * ktwilight_ Remote host closed the connection
1227020697 Q * ghislainocfs2 Quit: Leaving.
1227022561 Q * mtg Quit: Verlassend
1227023339 Q * chI6iT41 Ping timeout: 480 seconds
1227023439 J * ghislainocfs2 ~Ghislain@LPuteaux-151-41-11-129.w217-128.abo.wanadoo.fr
1227023441 J * mrfree ~mrfree@host1-89-static.40-88-b.business.telecomitalia.it
1227023457 M * mrfree hi all
1227023485 M * mrfree I noticed something "strange" about amavis-new and postfix in a guest
1227023497 M * mrfree and ip address
1227023519 M * mrfree both amavis and postfix bind 127.0.0.1
1227023607 M * mrfree but when postfix connects to the amavis daemon amavis (using net::server) detects 11.22.33.44 for the client ip but it is the eth0 IP addr of the HOST!!
1227023631 M * mrfree it should be "invisible" in the guest, isn't it?
1227023663 M * mrfree "ip addr" in the guest only reports "lo"
1227023699 M * mrfree any idea?
1227023723 M * mrfree I'm using linux-vserver 2.6.26-vs2.3.0.35.6
1227023724 M * PowerKe what do you mean by detects? Maybe you're using a dns name that's resolving to the external IP?
1227023742 M * PowerKe Or do you really see it on the incoming mail log as the remote IP?
1227023767 M * mrfree in the mail.log file amavis reports  DENIED ACCESS from 11.22.33.44
1227024025 M * mrfree I'm not sure but it could be an amavis-new issue
1227024039 M * mrfree because an old version on another guest works right
1227024059 M * PowerKe Did you grant any special capabilities to the guest?
1227024126 M * mrfree I cloned it (using rsync) from the working one, nothing more nothing less
1227024139 M * mrfree then I updated some software
1227024161 M * PowerKe ok, but nothing special in /etc/vservers/<guest>/bcapabilities ?
1227024170 J * dowdle ~dowdle@scott.coe.montana.edu
1227024200 Q * davidkarban Quit: Ex-Chat
1227024225 M * mrfree no, that file doesn't exist here
1227024338 M * PowerKe I'm running amavis-new on 2.2.0 and haven't encountered any similar problem
1227024378 M * mrfree I'm running amavis-new 2.6.1
1227024400 M * mrfree 2.5.2 seems to work fine
1227024416 M * PowerKe Same here: amavisd-new-2.6.1
1227024466 M * PowerKe I'm binding on localhost though, with localhost = 172.16.x.x in /etc/hosts
1227024526 Q * derjohn_mob Ping timeout: 480 seconds
1227024533 M * PowerKe You're sure the incoming connection is from postfix on the same guest and not from something on the host (or internet)?
1227024537 J * larsivi ~larsivi@9.80-202-30.nextgentel.com
1227024550 M * daniel_hozac mrfree: are you using SNAT?
1227024587 M * mrfree daniel_hozac yes
1227024618 M * daniel_hozac limited to your outbound interface?
1227024633 M * mrfree iptables -t nat -A PREROUTING -s 10.10.0.10 -p tcp -m multiport --destination-ports 25,993,636 -j DNAT --to-destination 192.168.0.7
1227024671 M * mrfree opss DNAT
1227024680 M * daniel_hozac no SNAT?
1227024749 M * mrfree iptables -t nat -A POSTROUTING -s 192.168.0.0/24 -d ! 192.168.0.0/24 -j SNAT --to-source $EXTIP
1227024766 M * daniel_hozac that's your problem.
1227024770 M * daniel_hozac add -o eth0
1227024985 M * mrfree daniel_hozac, I think you're right now the incoming connection comes from 192.168.0.7 that is the local guest IP
1227025029 M * mrfree which device should I use for /etc/vserver/.../interfaces/0/dev ??
1227025059 M * daniel_hozac the interface on which you want the IP address to be added.
1227025678 J * opuk ~kupo@potatisbulle.com
1227026040 J * doener_ ~doener@i577BAED4.versanet.de
1227026141 Q * doener Ping timeout: 480 seconds
1227026801 J * chI6iT41 ~chigital@tmo-100-193.customers.d1-online.com
1227027082 J * ntrs_ ~ntrs@77.29.17.126
1227027737 M * mrfree daniel_hozac, I noticed another little thing :) it I use  telnet localhost 10024 for example from the guest shell the ip that try to connects is 192.168.0.7 (eth0 in the guest) instead of 127.0.0.1
1227027763 M * daniel_hozac does /etc/hosts in the guest have 127.0.0.1 localhost in it?
1227027777 M * mrfree yes
1227027795 M * daniel_hozac do you have single_ip enabled for the network context?
1227027817 M * mrfree what do you mean with "single_ip"?
1227027850 M * mrfree actually ip addr in the guest reports lo 127.0.0.1 and eth0 192.168.0.7
1227028066 Q * chI6iT41 Ping timeout: 480 seconds
1227028067 M * mrfree daniel_hozac, I suppose using the shell I'm "using" 127.0.0.1
1227028426 M * mrfree daniel_hozac, ok I found what is the single_ip flag :) no I haven't enable any flag
1227028652 Q * ntrs_ Ping timeout: 480 seconds
1227029059 M * mrfree daniel_hozac, forcing ~SINGLE_IP worked :) thz
1227029546 Q * mrfree Quit: Leaving
1227030490 J * ktwilight ~ktwilight@87.66.193.174
1227030519 Q * Hurga Remote host closed the connection
1227031170 J * geb ~geb@4.4.82-79.rev.gaoland.net
1227031175 M * geb hi
1227031505 N * pmenier pmenier_off
1227032106 J * Slydder1 ~chuck@dslb-088-072-096-160.pools.arcor-ip.net
1227032212 J * cga ~weechat@94.36.130.212
1227032251 Q * Slydder1 
1227032289 J * Piet ~piet@86.59.118.153
1227032313 J * ntrs_ ~ntrs@77.29.17.126
1227033485 Q * kir Quit: Leaving.
1227033947 J * hparker ~hparker@2001:470:1f0f:32c:215:f2ff:fe60:79d4
1227033960 J * chI6iT41 ~chigital@tmo-096-120.customers.d1-online.com
1227034533 Q * chI6iT41 Ping timeout: 480 seconds
1227034588 Q * gnuk Quit: NoFeature
1227035124 J * chI6iT41 ~chigital@tmo-100-240.customers.d1-online.com
1227038178 J * dallas ~dallas@sf.newdream.net
1227039345 J * ntrs__ ~ntrs@77.29.11.119
1227039762 Q * ntrs_ Ping timeout: 480 seconds
1227040123 M * pflanze What is VSERVER_AUTO_LBACK exactly? The only page about it that I found in the wiki (http://linux-vserver.org/GroteblupsList) doesn't explain it.
1227040154 M * pflanze I guess I want "n" for the old behaviour?
1227040211 M * pflanze (I'm using aliases for creating ip's and iptable rules to define access)
1227040305 M * pflanze Hm. About same question for VSERVER_AUTO_SINGLE.
1227040310 M * pflanze Is this ngnet already?
1227040318 M * daniel_hozac AUTO_LBACK assigns a private loopback address to each guest.
1227040326 M * daniel_hozac no, mainline is doing ngnet.
1227040332 M * pflanze yes that's what the help text says
1227040340 M * micah pflanze: LBACK is very useful!
1227040351 M * pflanze micah: why?
1227040373 M * micah if you have any process that likes to bind to the loopback interface, this will be very useful
1227040373 M * pflanze I already assign a loopback ip to each guest.
1227040384 M * micah you no longer need to do that with LBACK
1227040403 M * pflanze Yes but then I'll have to rewrite all my setup stuff I guess
1227040411 M * pflanze I guess it doesn't buy me anything?
1227040421 M * pflanze If I'm already doing it with the config.
1227040432 M * daniel_hozac any service that really cares about 127.0.0.1 will be happy.
1227040447 M * pflanze But my guests already were happy with localhost services.
1227040448 M * micah except that each of your guests can see all the other guests over the loopback in your setup
1227040459 M * daniel_hozac there's that too.
1227040466 M * pflanze micah: nope, that's what the firewalling rules prevented
1227040472 M * micah with LBACK, you have a fully virtualized loopback
1227040479 M * micah pflanze: clean out the attic :)
1227040509 M * pflanze I'll only do that in the future when I know everything is working right.
1227040514 M * pflanze So I'm going with "n" for now.
1227040527 M * pflanze What about VSERVER_AUTO_SINGLE?
1227040534 M * daniel_hozac it does what the help text says.
1227040548 M * pflanze If I want the vs2.2 behaviour, I say "n", correct?
1227040558 M * daniel_hozac you say y.
1227040591 M * pflanze ehr. Then I don't understand it. Also, my guests all have 2 ip's.
1227040639 M * pflanze hm is this what chbind did?
1227040678 M * daniel_hozac what?
1227040716 M * daniel_hozac contexts with NXF_SINGLE_IP set will bind to the first IP, always.
1227040740 M * pflanze The chbind tool; it did set up the network contexts. iirc. hum.
1227040765 M * daniel_hozac AUTO_SINGLE sets NXF_SINGLE_IP automatically, and removes it when the context gets more than one IP.
1227040773 M * daniel_hozac just like it worked in vs2.2.
1227040799 M * pflanze What 2.2 did for me was: if some service binds to *, it gets the first ip. This is what I've relied on (the first ip is the "localhost" ip).
1227040804 M * pflanze k
1227040810 M * pflanze hm
1227040830 M * pflanze removes it. whatever, I guess if I have 2 ip's it's totally irrelevant anyway.
1227040845 M * daniel_hozac no, that's not what 2.2 does.
1227040857 M * daniel_hozac only if the context only has one IP will 0.0.0.0 bind to the first one.
1227040911 M * pflanze ( hm. Yes binding to * would bind to both ip's, true; what I've meant was that outgoing connections would use the first ip. )
1227040989 A * pflanze never heard of NXF_SINGLE_IP, anyway
1227041025 N * ensc Guest3922
1227041034 J * ensc ~irc-ensc@77.235.182.26
1227041136 Q * Guest3922 Ping timeout: 480 seconds
1227041275 Q * cga Quit: WeeChat 0.2.6
1227041281 M * pflanze Of course I wanted to say NGNet above. 
1227042261 Q * Piet Quit: Piet
1227042649 J * derjohn_mob ~aj@e180206070.adsl.alicedsl.de
1227044636 Q * Mojo1978 Remote host closed the connection
1227046096 Q * dna Quit: Verlassend
1227046458 Q * dallas Ping timeout: 480 seconds
1227047099 Q * chI6iT41 Ping timeout: 480 seconds
1227048004 J * chI6iT41 ~chigital@tmo-096-97.customers.d1-online.com
1227048579 Q * chI6iT41 Ping timeout: 480 seconds
1227048857 M * pflanze Hm, /proc isn't mounted anymore in a guest.
1227048890 M * pflanze in spite of the 'none    /proc           proc    defaults                0 0' entry in /etc/vservers/$vservername/fstab
1227048995 M * daniel_hozac what makes you say that?
1227049107 M * pflanze # lsof -i -n
1227049107 M * pflanze lsof: can't open /proc
1227049128 M * pflanze aha, # ls -lrt /proc/
1227049128 M * pflanze ls: /proc/: Permission denied
1227049151 M * pflanze hm. why is this?
1227049184 J * chI6iT41 ~chigital@tmo-100-129.customers.d1-online.com
1227049200 M * daniel_hozac what kernel?
1227049209 M * pflanze 2.6.27.6-vs2.3.0.35.10
1227049254 M * pflanze previously I was running 2.6.22.x + vs2.2
1227049376 M * daniel_hozac setattr --~hide /proc
1227049430 M * pflanze on the host? on which dir, the host's /proc ?
1227049452 M * pflanze showattr /proc | head -1 is
1227049453 Q * chI6iT41 Read error: Connection reset by peer
1227049455 M * pflanze AwH-bui- /proc/
1227049472 A * pflanze tries
1227049537 M * pflanze that helps; but hm is it safe?
1227049555 M * daniel_hozac safe to give the guest access to the /proc root directory?
1227049561 M * daniel_hozac of course.
1227049595 M * pflanze I mean, it sounds a bit strange to set 'permissions' on the *host's* proc so that the guest can use it
1227049607 M * daniel_hozac there's just one /proc.
1227049628 M * pflanze What's the purpose of the hide flag?
1227049666 M * pflanze and why isn't it cleared by vproc_unhide or so?
1227049711 M * daniel_hozac it's a bug.
1227049719 M * pflanze k
1227049768 Q * eyck Ping timeout: 480 seconds
1227050573 M * pflanze Thanks, things seem to be working. This is a test machine, tell me if you'd like me to test something in particular.
1227051231 J * mugwump ~samv@watts.utsl.gen.nz
1227051251 M * mugwump anyone know how to force 'vserver xxx exec' to allocate a psuedo-terminal?
1227051264 M * mugwump this used to happen with util-vserver 0.30.210, but not 0.30.212
1227051282 M * mugwump at least I'm trying to figure out why I can't 'vserver xxx exec login -f username'
1227051446 A * mugwump will check back later &
1227051450 J * eyck ocqRlemd@nat06.nowanet.pl
1227051477 Q * hparker Quit: Quit
1227051560 M * pflanze I can confirm that exec login -f user exits right away; but "vserver foo exec bash -c 'lsof -p $$'" *does* show /dev/pts/$x as stdin/out so?
1227051634 M * pflanze well, the kernel says (only with the login, not the lsof): vxW: [»login«,11439:#1001|1001|1001] did lookup hidden devpts:f5bc157c[#0,2] »/dev/pts«.
1227051668 M * pflanze (and the vserver tool exits with status code 1)
1227051780 M * daniel_hozac mugwump: enter is the only one that allocates a tty.