1166314515 J * DreamerC ~dreamerc@125-225-100-47.dynamic.hinet.net 1166315261 Q * dna Quit: Verlassend 1166315599 Q * bonbons Quit: Leaving 1166316855 N * Bertl_oO Bertl 1166319011 Q * Aiken Quit: Leaving 1166320115 Q * Piet_ Quit: Piet_ 1166321491 Q * meandtheshell Quit: Leaving. 1166325560 Q * Zaki Ping timeout: 480 seconds 1166326693 J * Zaki ~Zaki@88.213.57.72 1166328475 Q * Zaki Ping timeout: 480 seconds 1166329761 J * Aiken ~james@tooax6-004.dialup.optusnet.com.au 1166329924 M * Bertl wb Aiken! 1166330043 M * Aiken hi 1166330059 M * Bertl Aiken: btw, the lback stuff is not in the 2.3.x release yet 1166330092 M * Bertl Aiken: seems I removed it on the 2.6.19 transition, but it will be back in the next 2.3.x release 1166330152 M * Aiken with 2.3.0.4 on the host and 2 guests each could ssh user@localhost and connect to localhost 1166330167 M * Aiken what will lback provide over that? 1166330207 M * Bertl I assume the 2.3.0.4 worked as the 2.2.x would 1166330219 M * Bertl i.e. it remapped the 127.0.0.1 to the first assigned ip 1166330232 M * Bertl (which is a config option) 1166330234 M * Aiken ok 1166330245 M * Bertl the next 2.3.x will actually use the lback code 1166330260 M * Bertl so binding to 127.x.y.1 will look like binding to 127.0.0.1 inside 1166330271 M * Aiken any chance you know how to migrate a bugzilla with mysql backend to bugzilla with a postgresql back end? 1166330290 M * Aiken that should be for anything that explicitly wants 127.0.0.1 1166330344 M * Bertl yep, precisely 1166330364 M * Aiken I did not encounter any problems yesterday. I was thinking of upgrading this machine to 2.3.x as well. 1166330365 M * Bertl regarding the bugzilla migration, I'd think you want to copy the tables 1166330371 M * Aiken it is still 2.6.18-vs2.1.1-rc38 1166330390 M * Bertl probably 2.2.0-rc5 is what you want atm 1166330406 M * Bertl should have almost all features, and is pretty stable by now 1166330422 M * Bertl 2.3.x will be interesting for testing, as it will get all the new network code 1166330644 M * Aiken updated ipv4 or mainly introducing ipv6? 1166330661 M * Bertl both 1166330681 M * Bertl better abstraction for ipv4, similar for ipv6 1166330702 M * Bertl most restrictions on the ip assignment removed 1166330721 M * Aiken have not done anything with ipv6 yet 1166330741 M * Aiken most I have done is a quick look to see what ipv6 tunnel brokers are about 1166330749 M * Bertl except for reading and testing a few things, me neither 1166334807 J * FireEgl Sebastian@adsl-61-147-76.bhm.bellsouth.net 1166335730 J * Zaki ~Zaki@88.213.43.1 1166339390 M * Bertl okay, off to bed now ... have a good one everyone! 1166339395 N * Bertl Bertl_zZ 1166340852 Q * ensc Ping timeout: 480 seconds 1166341346 Q * dmax Ping timeout: 480 seconds 1166341362 Q * s0undt3ch Ping timeout: 480 seconds 1166341453 J * ensc ~irc-ensc@p54B4D820.dip.t-dialin.net 1166344252 P * akaihola 1166346316 J * dna ~naucki@161-236-dsl.kielnet.net 1166350228 M * yang daniel_hozac: I am getting error vc_net_create http://paste.linux-vserver.org/783 1166350540 J * bonbons ~bonbons@83.222.37.103 1166350677 M * yang hey bonbons 1166350697 M * bonbons hey yang 1166350719 M * yang do you know why i am getting this error 1166350725 M * yang vc_net_create http://paste.linux-vserver.org/783 1166350739 M * yang i got 0.30.212 now 1166351040 M * bonbons from the output I can't tell... what content do you have in /etc/vservers/hek/interfaces/* ? 1166351061 J * akaihola ~akaihola@dsl-aur-fecff800-7.dhcp.inet.fi 1166351184 M * akaihola The SHA-1 algorithm used by vhashify doesn't seem to generate hashes identical to those by sha1sum from the coreutils package. I wonder why. 1166351503 M * bonbons yang: if you can get the complete command line for ncontext that fails it would be helpful for determining the cause 1166351826 M * yang [11:36:03] root@jason /etc/vservers/hek/interfaces # ls 1166351826 M * yang 0 1 10 11 12 13 14 15 2 3 4 5 6 7 8 9 1166351826 M * yang cd 0 ; cat * eth0, 86.110.64.160, 24 1166352027 M * bonbons all are IPv4? 1166352044 M * yang all 1166352049 M * yang only ipv4 1166352072 J * meandtheshell ~markus@85-124-174-231.dynamic.xdsl-line.inode.at 1166352142 Q * Aiken Ping timeout: 480 seconds 1166352335 M * bonbons yang: could you add some debug echo to chbind script? 1166352360 Q * meandtheshell 1166352376 J * meandtheshell ~markus@85-124-174-231.dynamic.xdsl-line.inode.at 1166352463 M * yang bonbons: can you tell me how? 1166352553 M * bonbons looks like that one gets installed to /usr/sbin/chbind, so open your favorite editor and duplicate line 125 (near to the end in a if bloc) 1166352587 Q * ruskie Quit: Caught sigterm, terminating... 1166352600 M * bonbons line is "${create_cmd[@]}" "${chain_cmd[@]}" -- "$@" 1166352621 M * bonbons and put an echo in front of the copy 1166352644 M * bonbons but put the copy first, otherwise the return value will not be handled correctly 1166352768 M * yang echo "${create_cmd[@]}" "${chain_cmd[@]}" -- "$@" 1166352768 M * yang "${create_cmd[@]}" "${chain_cmd[@]}" -- "$@ 1166352771 M * yang like this? 1166352817 M * bonbons yes 1166352840 M * bonbons then start the guest again and check the output 1166352870 M * yang http://paste.linux-vserver.org/784 1166353071 M * bonbons fine, with that we can start figuring out... let's check the source of ncontext 1166353143 M * bonbons yang: are you using dynamic context IDs? 1166353179 M * yang hm, dynamic ? i don't know 1166353203 M * bonbons if so, you should switch to static ones... /etc/vserver/tek/context should contain the context ID 1166353270 M * yang there is no /etc/vservers/tek dir 1166353341 M * bonbons hek, sorry for miss-naming your guest 1166353378 M * yang you mean /etc/vservers/hek/context 1166353384 M * bonbons yes 1166353386 M * yang there is no context there 1166353392 M * yang shall i create one manually? 1166353431 M * bonbons then you are using dynamic context IDs... you should really put a number there (between 2 and 30000) 1166353437 M * yang ok 1166353519 M * bonbons dynamic context IDs are deprecated and support will be removed from kernel (if it isn't already done for some versions) 1166353557 M * yang OK great, it started now ! :) 1166353613 M * yang and for IPv6 it's grabbing ipv6 interfaces from root-server? 1166353644 M * yang if they are to be assigned there, they will work for guests 1166353645 M * bonbons it's grabbing interfaces you specify somewhere below interfaces/ 1166353673 M * yang bonbons: thank you ! 1166353702 M * bonbons the mode of operation should be the same as for ipv4, but I haven't used util-vserver-0.30.212 yet, so it's still assumption 1166353913 J * MaM ~ometra@mc-media.com 1166354354 M * Borg- frfrf/w 4 1166356231 J * dmax ~semaj@81.193.63.60 1166356258 M * matti :) 1166356322 J * s0undt3ch ~s0undt3ch@bl4-63-60.dsl.telepac.pt 1166356891 Q * ensc Killed (NickServ (GHOST command used by ensc_)) 1166356901 J * ensc ~irc-ensc@p54B4E0AE.dip.t-dialin.net 1166357756 M * bonbons Hollow: libvserver, src/lib/vserver.h@r235, line:594; are you sure the #ifndef _LINUX_SCHED_H is correct? 1166357913 M * Hollow bonbons: yep.. 1166357928 M * Hollow probably those for /usr/include/sched.h should be checked as well 1166357966 M * bonbons on line 594 it's for namespace (related) things... 1166358093 M * Hollow yeah .. 1166358239 M * bonbons you did change a lot in vserver.h since august/september ;) nearly everything got renamed! 1166358757 Q * MaM Ping timeout: 480 seconds 1166358932 Q * Rich_Estill Ping timeout: 480 seconds 1166358975 Q * tso Ping timeout: 480 seconds 1166359903 M * waldi Hollow: hmm, underscore and uppercase letter is reserved namespace. this statement emits undefined behaviour 1166361886 M * Johnnie Is there a changelog somewhere of the changes from 2.1.1 to 2.3.0.3? 1166362128 Q * Johnnie Read error: Connection reset by peer 1166362935 Q * FireEgl Quit: ... 1166362960 J * Johnnie ~jdlewis@jdlewis.org 1166364256 J * vittorio ~chatzilla@adsl-182-159.38-151.net24.it 1166364317 M * vittorio anybody knows how to apply Linux-VServer patch ion 2.6.17.13 kernel vs2.0.2.1 with suspend2? 1166364383 Q * vittorio 1166365381 J * lilalinux ~plasma@80.69.41.2 1166366470 Q * michal` Ping timeout: 480 seconds 1166366629 J * barnesa ~barnesa@prelude.ramsesit.com 1166366631 M * barnesa hi guys 1166367027 J * michal` ~michal@www.rsbac.org 1166367027 Q * Johnnie Read error: Connection reset by peer 1166367929 J * Johnnie ~jdlewis@jdlewis.org 1166374417 M * bonbons Hollow: did you run libvserver with latest kernels? (2.6.19 based), for me it always fails entering namespace ... 1166374610 M * Hollow bonbons: no, not yet.. just updated the headers and defines, but didn't test it yet 1166374705 M * bonbons ok, so will have to find out where it fails exactly (possibly related to context creation)... (Invalid argument) and libvserver dos not allow any argument! 1166375106 M * bonbons hmm, looks like the kernel does not accept older version calls anymore ... 1166375546 M * bonbons Bertl_zZ: vs2.1.1.7 causes troubles with vc_enter_space() ... "old" syscalls always fail with "Invalid Argument" 1166375779 M * bonbons shouldn't vc_enter_space() set mask to 0 instead of ~0 is data == NULL? 1166376501 M * AndrewLee pcman: sata 的 NCQ support 得用 ahci 才有支援 1166376517 M * AndrewLee oops, sorry, wrong channel. 1166376718 J * FireEgl ~FireEgl@adsl-61-147-76.bhm.bellsouth.net 1166377105 Q * Johnnie Quit: G'bye! 1166377379 M * daniel_hozac bonbons: you mean it doesn't? 1166377443 M * daniel_hozac ah yes, enter is broken. 1166377979 N * Bertl_zZ Bertl 1166377992 M * Bertl morning folks! 1166378164 M * Bertl ah, so that probably is the problem older tools encountered 1166378200 M * Bertl bonbons: did you find the new 2.3 release I uploaded for you? 1166378292 M * daniel_hozac morning Bertl! 1166378357 M * bonbons morning Bertl! 1166378370 M * bonbons did not check for 2.3 today, I was doing work on my /proc/cmdline patch and upgrading the vserver-utils to HEAD of libvserver 1166378371 M * daniel_hozac bonbons: re: using VCMD_ctx_create_v2, it'd require quite a bit of work. 1166378387 M * Bertl bonbons: cmdline patch? 1166378407 M * daniel_hozac (most of it is the same thing required to upgrade to _v1 :)) 1166378426 M * bonbons yep, a small patch to make per-guest content for /proc/cmdline 1166378439 M * Bertl what's that for? 1166378446 M * bonbons daniel_hozac: hehe 1166378513 M * Bertl daniel_hozac, bonbons: has anybody tested the .mask=0 change, if it fixes the issue? 1166378515 M * bonbons it's to allow giving guest's init-system some extra information (like the runlevel=xxxx you can provide for full hosts), that way guest's can start into a different runlevel 1166378538 M * Bertl i.c. is the patch uploaded somewhere? 1166378542 M * bonbons not tried yet to kill the '~' char in there 1166378542 M * daniel_hozac bonbons: hmm, util-vserver already allows that. 1166378564 M * bonbons not yet, but will put it to people.linux-vserver.org soon 1166378572 M * daniel_hozac (passed as an argument to /sbin/init) 1166378575 M * Bertl daniel_hozac: probably not for plain init + strange scripts parsing /proc/cmdline :) 1166378591 M * daniel_hozac Bertl: but just plain init does it fine :) 1166378603 M * Bertl yeah, I know 1166378612 M * bonbons daniel_hozac: but argument to /sbin/init is not sufficient, that one is nearly never visible to rest of boot sequence 1166378639 M * daniel_hozac bonbons: hmm? surely things are using /sbin/runlevel to figure out what runlevel it is? 1166378646 M * daniel_hozac which should be populated. 1166378769 M * bonbons e.g. gentoo does check for runlevel=xxx name, it's not using the runlevel numbers, in addition some boot sequences (with autodetection) allow many options on kernel command line to disable/enable operations 1166378944 M * daniel_hozac so gentoo doesn't parse init's arguments? 1166379562 M * bonbons gentoo gets started by the arguments to /sbin/rc in inittab, but more information is taken right from /proc/cmdline, especially the softlevel which determines what services/daemons get started 1166379614 M * Bertl anyway, I consider that feature useful, when properly implemented 1166379793 Q * yarihm Quit: This computer has gone to sleep 1166380056 M * bonbons Bertl: setting default mask to 0 in vc_enter_space() instead of ~0 works at least 1166380131 M * Bertl okay, looks fine to me after the enter changes we did 1166380506 M * bonbons Bertl: http://people.linux-vserver.org/~bonbons/patches/delta-cmdline-feat03.diff 1166380633 M * Bertl hmm, why not extend the vhi interface? 1166380677 M * Bertl (i.e. I don't like the hacked create interface) 1166380679 M * daniel_hozac that's what i was thinking too. 1166380719 M * bonbons the reason I put it into create is that /proc/cmdline is expected to remain constant once guest is created 1166380726 M * Bertl vhi could benefit from a new 'posix' conform extension to 256 chars ... 1166380740 M * Bertl bonbons: so are most vhi entries 1166380763 M * Bertl that you _can_ change them doesn't mean that you actually change them 1166380843 M * bonbons in addition the vhi part looks like being the new uts-namespace which /proc/cmdline is not part of 1166380872 M * Bertl neither is the context name 1166380890 M * bonbons true 1166380960 M * Bertl sidenote: we have a set of reserved 'test' categories, which should be used for 'testing' stuff in patches 1166380980 M * Bertl otherwise we'll end up with broken ABI 1166381140 M * bonbons is that one VC_CAT_SYSTEST? 1166381205 M * Bertl that is one, but the entire column is reserved for testing 1166381214 M * Bertl see include/linux/vserver/switch.h 1166381243 M * Bertl but yes the cat 61 was traditionally used 1166382602 M * bronson Bertl: glanced at 2.6.20 yet? Just wondering if you have any impressions. 1166382634 J * Johnnie ~jdlewis@jdlewis.org 1166383013 J * ruskie ~ruskie@ruskie.user.oftc.net 1166383113 M * Bertl bronson: nope, but will do so shortly 1166383129 M * bronson cool. No hurry, I'm just very curious what you find out. 1166383527 Q * Johnnie Read error: Connection reset by peer 1166384217 M * barnesa quit 1166384218 Q * barnesa Quit: Leaving 1166384315 J * Johnnie ~jdlewis@jdlewis.org 1166385109 J * dreamind ~dreamind@C2107.campino.wh.tu-darmstadt.de 1166385151 M * Bertl wb Johnnie! dreamind! 1166385160 M * dreamind Hi Bertl :D 1166385164 M * dreamind hi folks ;) 1166385190 J * Smutje ~Smutje@xdsl-87-78-98-134.netcologne.de 1166385805 M * Johnnie Hey Bertl. 1166385819 M * Bertl wb Smutje! 1166386656 M * nebuchadnezzar for my problem about hashify which do not work, http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=403332 1166386676 M * nebuchadnezzar Ola Lundqvist think it's a kernel problem :-/ 1166386724 M * nebuchadnezzar arf the bug page is not up to date 1166386742 M * Bertl nebuchadnezzar: well, Ola thinks a lot :) 1166386781 M * nebuchadnezzar I have some "operation not supported" during a hashifu 1166386785 M * nebuchadnezzar hashify 1166386799 M * Bertl okay, I'm not saying he is wrong, let's do some test, shall we? 1166386806 M * nebuchadnezzar ok 1166386817 M * Bertl you have a shell on that machine? 1166386835 M * Bertl (not for me, for you :) 1166386837 M * nebuchadnezzar yes 1166386854 M * Bertl okay, do you have an unused partition there too? 1166386886 M * Bertl if not, create a file somewhere like this 1166386891 M * nebuchadnezzar I can create one on my LVM 1166386908 M * Bertl okay, then do that, roughly 200MB 1166386923 M * Bertl it will be overwritten in the test process 1166386925 M * nebuchadnezzar http://www.asgardr.info/~nebu/vserver/ola.mbox <-- for Ola response 1166386967 M * nebuchadnezzar ok 200Mb logical volume created 1166386972 M * nebuchadnezzar 200MB 1166386988 M * Bertl excellent, now get the testfs.sh script from here: 1166387023 M * Bertl http://vserver.13thfloor.at/Stuff/SCRIPT/testfs.sh 1166387039 M * Bertl make it executable, and do the following: 1166387055 M * Bertl - create a directory as mountpoint (mkdir /test) 1166387066 M * nebuchadnezzar ok 1166387115 M * Bertl - run 'testfs.sh -tx -D 1166387125 M * Bertl (adjust to be your lvm 1166387142 M * Bertl - upload the output to paste.linux-vserver.org 1166387147 M * hardwire` meh 1166387151 M * hardwire` who here is having fun today! 1166387162 A * Bertl raises his hand 1166387171 M * hardwire` you lie 1166387195 M * nebuchadnezzar http://paste.linux-vserver.org/785 1166387213 M * Bertl hardwire`: I never lie 1166387221 M * hardwire` like an elephant!? 1166387241 M * Bertl nah, they never forget, you got that one wrong :) 1166387269 M * hardwire` you lie! 1166387270 M * Bertl nebuchadnezzar: so it seems the kernel is working quite fine (the 116 test failing is expected with CoW link breaking) 1166387322 M * Bertl daniel_hozac: any known issues with CoW and vhashify? 1166387351 M * hardwire` Bertl: you the man with cow usage? 1166387366 M * hardwire` its either cow via vserver or cow via ZFS 1166387375 M * hardwire` or nilfs 1166387382 M * Bertl hardwire`: hmm, well, different layer 1166387402 M * hardwire` Bertl: is it referenced/howto'd on linux-vserver.org? 1166387408 M * nebuchadnezzar Bertl: during the test I have some messages in dmesg 1166387427 M * Bertl nebuchadnezzar: those are fine, they are part of the tests 1166387445 M * Bertl hardwire`: I think we do not have a separate page on CoW yet 1166387450 M * nebuchadnezzar the denied access, etc... 1166387494 M * hardwire` so vhashify merges like files in two trees eh? 1166387500 M * hardwire` via hard links. 1166387508 M * Bertl yes, plus some extra flags 1166387510 M * hardwire` and offers no copy-on-write itself for changes to hard linked files 1166387518 M * Bertl (to allow for the CoW link breaking) 1166387520 M * hardwire` like breaking a hard link 1166387523 M * hardwire` ahha! 1166387524 M * hardwire` I am smart 1166387544 M * hardwire` so basically its more like a break-the-link layer :) 1166387562 M * Bertl yes, except for the fact that it isn't a separate layer 1166387585 M * Bertl it works within the normal vfs and is fairly filesystem independant 1166387590 M * hardwire` ok 1166387600 M * hardwire` so a vserver has some control over how it stores files 1166387628 M * hardwire` kernel level.. pre write? atleast? 1166387652 M * Bertl nebuchadnezzar: have you tried with a mainline (not debian) dietlibc from cvs and a mainline 0.30.212? 1166387668 M * nebuchadnezzar no 1166387683 M * Bertl (probably same goes for beecrypt 1166387707 M * Bertl I would give that a try, and see if it fixes the issue for you 1166387738 M * nebuchadnezzar I just need the util-vserver tarball ? 1166387765 M * Bertl yes 1166387781 M * Bertl but make sure that the dietlibc is built and installed first 1166387794 M * Bertl otherwise it will be built with the old/wrong diet 1166387880 M * nebuchadnezzar I haven't dietlibc installed 1166387910 M * nebuchadnezzar arf, just required for compilation... 1166387974 M * Bertl make sure you get 0.30 or later 1166388207 M * nebuchadnezzar ok, dietlibc is compiling 1166388524 Q * Johnnie Remote host closed the connection 1166388762 M * hardwire` I imagine it would 1166388930 M * hardwire` is the networking proxy arped? 1166388949 M * hardwire` nm 1166388964 M * hardwire` I would have to proxy arp it if I associated it to a different ethernet device 1166389306 J * Johnnie ~jdlewis@jdlewis.org 1166389644 J * Aiken ~james@tooax6-247.dialup.optusnet.com.au 1166389664 M * Bertl morning Aiken! 1166389684 M * Aiken hi 1166389924 M * Aiken anyone else getting this type of message from the list server? http://paste.linux-vserver.org/786 1166390051 M * Bertl yes, got that once after a broken spam blacklist 1166390059 Q * FireEgl Remote host closed the connection 1166390072 M * Bertl you have to logon to the mailing list manager, there you can reenable it 1166391060 M * nebuchadnezzar 'exec-ulimit: execv(): Bad address' 1166391070 M * nebuchadnezzar erf 1166391219 M * nebuchadnezzar time to sleep now, I'll see that tomorrow 1166391250 M * Bertl that means broken dietlibc 1166391268 M * nebuchadnezzar it's a fresh compile one 1166391269 M * Bertl it was fixed on ppc with the cvs version 1166391277 Q * meandtheshell Quit: Leaving. 1166391278 M * Bertl what dietlibc did you build? 1166391285 M * nebuchadnezzar 0.30 1166391295 M * nebuchadnezzar from fefe 1166391303 M * Bertl okay, what distro are you on? 1166391307 M * nebuchadnezzar debian 1166391317 M * nebuchadnezzar sid 1166391321 M * Bertl hmm, okay, get the version from cvs (tomorrow) and try again 1166391334 M * nebuchadnezzar ok 1166391336 M * nebuchadnezzar see you 1166391342 M * Bertl if that fails too, try the latest debian version, maybe they add some patches 1166391356 M * Bertl cya 1166391508 M * bonbons Bertl: shouldn't strings be forced to terminate with '\0' in vhi*? 1166391535 Q * dreamind Quit: dreamind 1166391536 M * Bertl not require with the uts interfaces, IIRC 1166391586 M * bonbons sure that information is never used as \0 delimited string (e.g. in printfs)? 1166391612 M * Bertl can't hurt to verify that 1166391742 M * Bertl at least cvirt_proc does the right thing 1166391841 M * Bertl and vx_name is not exposed in the procfs 1166391862 M * Bertl (only via syscall interface, which is fine) 1166391951 M * bonbons I've seen for cvirt, need to see if uts information is also used that way in mainline 1166392101 M * bonbons I see things like strlen(nodename) ... that fill cause trouble if not \0 terminated... 1166392325 M * bonbons e.g. kernel/sys.c:1825 sys_gethostname() 1166392511 M * Bertl indeed, the mainline kernel portions depend on 0 termination 1166393224 M * Bertl bonbons: so basically the userspace interface should better be careful and zero terminate the string atm 1166393276 M * bonbons yep, I guess for all strings it would be careful to always 0-terminate, unless the length is passed with it 1166393391 M * bonbons but when it's not 0 terminated a warning comment should be present... 1166393495 M * bonbons that way we prevent risk of buffer-overflows even when code gets updates at a later time (and for data shared with mainline a mainline change might slip past our attention very easily) 1166393831 M * Bertl I think it would be a sane thing to force the last byte to 0 1166393847 M * Bertl (on copying from userspace in the vhi functions) 1166393926 M * Bertl daniel_hozac: where would I find the FC packages for util-vserver? 1166394031 J * yarihm ~yarihm@84-75-123-221.dclient.hispeed.ch 1166394173 M * bonbons Bertl: removal of legacy and reorganization of network code in 2.3.0.5 look intresting... but I wonder why you put nx_dev_visible() in vs_inet.h ... 1166394219 Q * akaihola Quit: Leaving. 1166394225 M * bonbons I would rather see it in a stack-independent header, so that it calls the right functions on ipv4, ipv6 or other stacks if they ever happended to be supported 1166394300 M * bonbons there are probably other such functions which should be treated the same way 1166394573 M * Bertl yes, I have no problem with that, it was the first cleanup 1166394595 M * Bertl we could as well put it into vs_network 1166394616 M * bonbons yep 1166394625 M * Bertl but you probably want stack specific versions of that anyway 1166394642 M * Bertl i.e. you want a 'wrapper' called nx_dev_visible() 1166394651 M * Bertl and check functions per stack 1166394659 M * bonbons exactly 1166394705 M * Bertl btw, do you think it is realistic to handle ipv6 addresses which are ipv4 compat mappings with the ipv4 checks? 1166394749 M * bonbons I guess ipv6 will have to call ipv4 functions for mapped addresses 1166394763 M * Bertl okay, that is what I meant 1166394787 M * Bertl I would prefer to handle it that way, i.e. extract the ipv4 part and pass it to the ipv4 checks 1166394807 M * bonbons that's the cleanest way 1166394840 M * bonbons and those calls are required only if dual-stack sockets are allowed 1166394915 Q * michal` Ping timeout: 480 seconds 1166394951 M * bonbons will look at it more in depth tomorrow 1166394955 M * Bertl okay, tx 1166394967 A * bonbons goes sleeping 1166394975 M * Bertl have a good one then! 1166394992 M * bonbons thanks, and good coding! 1166395028 Q * bonbons Quit: Leaving 1166395223 J * michal` ~michal@www.rsbac.org 1166395261 M * Bertl wb michal`! 1166396891 Q * yarihm Quit: This computer has gone to sleep 1166399312 Q * matti Ping timeout: 480 seconds