1181697200 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181697200 T * daniel_hozac - 1181697206 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181697206 T * daniel_hozac - 1181697219 M * Bertl hmm .. strange ... 1181697362 M * coderanger_ Hmm, ctypes isn't liking libvserver.so 1181697392 M * Bertl how so? 1181697421 M * coderanger_ Not finding any function symbols that I have tried so far 1181697429 A * coderanger_ checks his compiler arguments 1181697807 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181697807 T * daniel_hozac - 1181697955 M * coderanger_ Bertl: Am I correct in guessing the ".213" means revision 213? 1181697973 M * coderanger_ Bertl: As the libvserver API seems to differ from /trunk 1181697983 M * Bertl yeah, but it is a little trickier with the numbering 1181698030 M * coderanger_ Are we better off with the version in the download, or should be go for the latest libvserver (and therefore the new C API)? 1181698077 M * Bertl won't hurt to use the latest, but it shouldn't differ much from .213 1181698086 M * coderanger_ Its a pretty big diff on vserver.h 1181698095 M * Bertl really? 1181698109 M * coderanger_ http://svn.linux-vserver.org/projects/libvserver/changeset?new=trunk%2Fsrc%2Flib%2Fvserver.h%40246&old=trunk%2Fsrc%2Flib%2Fvserver.h%40211 1181698122 M * coderanger_ Maybe not as big as you are used to dealing with ;-) 1181698166 M * coderanger_ But I'll just grab libvserver from trunk and build that 1181698217 M * Bertl make that 1181698318 M * coderanger_ roger 1181698328 J * slacker403 ~root@cpe-71-74-73-121.insight.res.rr.com 1181698337 M * Bertl wb slacker403! 1181698379 M * neuralis Bertl: we're now ready to produce a kernel rpm with vs when we get your updated patch, and start testing immediately 1181698388 M * slacker403 Bertl: just been compiling stuff 1181698417 M * Bertl neuralis: okay, will take at least another day of cleaning up ... 1181698422 M * neuralis no problem 1181698431 M * slacker403 Bertl: i guess i will patch for Vserver now 1181698435 M * Bertl neuralis: but good to know ... 1181698438 M * neuralis that we've finally gotten there is important :) 1181698440 A * neuralis & 1181698503 M * Bertl neuralis: ah, btw, we need ext2/3 + jffs2 support, but we could remove the other filesystem, unless you want to use them on the future server or so? 1181698581 M * Bertl neuralis: I mean, the additional attribute and tagging support ... 1181698777 M * coderanger_ Looks like ./configure is looking for a libucid, anyone know what that is? 1181698814 M * coderanger_ It looks like it is part of VCD, but we weren't planning to use that 1181698962 M * coderanger_ Hmm, the source SVN server for it is down 1181698966 M * coderanger_ Anyone have a copy? 1181698989 M * Bertl http://people.linux-vserver.org/~hollow/ 1181699011 M * coderanger_ Gratsi 1181699022 M * Bertl np 1181699187 M * coderanger_ Hmm, now the load paths dance 1181699486 M * coderanger_ Hollow: Ping 1181699495 J * Ashsong ~mstone@pepper.cs.swarthmore.edu 1181699514 M * coderanger_ It seems to not be converting the syscall.m4 file 1181699527 M * Bertl welcome Ashsong! 1181699545 M * coderanger_ (and then complaining when it can't find _NR_vserver and such) 1181699546 M * Ashsong Bertl: Hello. I'm working with coderanger_. 1181699568 M * Bertl good to know :) 1181699734 M * Ashsong Bertl: It turns out that coderanger_ effectively ran "autoreconf" after SVN checkout, rather than using the provided (but easy to miss) Makefile.svn 1181699778 M * Ashsong Bertl: Consequently, autoconf was called without the "-I m4" flag and missed the syscall definitions. 1181699784 M * coderanger_ Now it wants a non-existent lucid/mem.h 1181699799 M * coderanger_ Sounds like a version mismatch w.r.t. lucid 1181699954 A * coderanger_ retries with trunk lucid 1181700126 M * neuralis Bertl: i don't think we'll use anything more exotic than ext3 on the server 1181700274 M * coderanger_ neuralis: Come on, itl be %0.0000001 faster on Reiser4! 1181701150 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181701150 T * daniel_hozac - 1181701150 M * Bertl ... 1181701150 M * coderanger_ Bertl: It seems like lucid is unhappy when it tries to find my ARCH 1181701184 M * Bertl what ARCH do you have? 1181701211 M * coderanger_ neuralis: What is in live? I had thought it was an x86 1181701225 M * coderanger_ neuralis: It is detecting as ARCH=1 1181701762 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181701762 T * daniel_hozac - 1181702180 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181702180 T * daniel_hozac - 1181702261 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181702261 T * daniel_hozac - 1181702289 M * Bertl coderanger_: pong? 1181702298 M * coderanger_ Can you run this: echo "i386" | cpp 1181702314 M * coderanger_ For some reason that spits out one, and I have no idea why 1181702352 M * Bertl probably because i386 is defined on your machine? 1181702397 M * coderanger_ I thought the cpp define for it was __i386__ 1181702402 M * Bertl try echo | cpp -dM 1181702428 M * coderanger_ Yeah, I see a #def i386 1 1181702441 M * Bertl mystery solved :) 1181702442 M * coderanger_ Is there a good way to trace it? 1181702454 M * Bertl trace it? 1181702475 M * coderanger_ As in to know why it is doing that, as it breaks the lucid build scripts 1181702509 M * Bertl I guess it is compiled into the gcc or configured in the cpp macros? 1181702593 M * coderanger_ Ugh, I'll just hardcode it for now 1181702633 M * coderanger_ Is there a bug tracker for lucid somewhere that I should report this? 1181702671 M * Bertl probably best to report it to Hollow directly 1181702694 M * coderanger_ Hollow: Your arch detection is teh b0rked ;-) 1181703167 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181703167 T * daniel_hozac - 1181704109 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181704110 T * daniel_hozac - 1181704110 J * Aiken_ ~james@ppp121-45-206-37.lns1.bne1.internode.on.net 1181704110 Q * Aiken Ping timeout: 600 seconds 1181704192 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181704192 T * daniel_hozac - 1181704433 Q * Ashsong Quit: leaving 1181704624 J * Aiken ~james@ppp121-45-192-48.lns1.bne1.internode.on.net 1181704679 Q * Aiken_ Ping timeout: 600 seconds 1181704739 Q * coderanger_ Ping timeout: 600 seconds 1181705102 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181705102 T * daniel_hozac - 1181705402 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181705402 T * daniel_hozac - 1181705703 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181705703 T * daniel_hozac - 1181706003 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181706003 T * daniel_hozac - 1181708183 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181708183 T * daniel_hozac - 1181708838 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181708838 T * daniel_hozac - 1181709078 M * Bertl ... 1181709552 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181709552 T * daniel_hozac - 1181709907 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181709907 T * daniel_hozac - 1181709937 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181709937 T * daniel_hozac - 1181709971 M * Bertl ... 1181711976 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181711976 T * daniel_hozac - 1181712033 M * Bertl ... 1181713765 T * * http://linux-vserver.org/ | latest stable 2.2.0, 2.0.3-rc2, devel 2.3.0.12, stable+grsec 2.0.2.1, 2.2.0 | util-vserver-0.30.213 | 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 ;) 1181713765 T * daniel_hozac - 1181713792 M * Bertl okay ... (waiting for 10min silence :) 1181713986 J * sharkjaw ~gab@158.36.45.236 1181713994 M * Bertl welcome sharkjaw! 1181714007 M * slacker403 Bertl: i was jus gonna use the newest one from kerrnel.org 1181714018 M * Bertl 2.6.22-rc4? 1181714021 M * slacker403 2.6.21.5 1181714033 M * slacker403 latest stable 1181714120 M * Bertl http://vserver.13thfloor.at/Experimental/patch-2.6.21.5-vs2.2.0-rc1.diff 1181714152 M * daniel_hozac new patch for 21.5? does that have the pid_task fix? 1181714324 M * Bertl daniel_hozac: maybe :) 1181714349 M * Bertl I remember adding that one to my tree some time ago, but I'm not 100% sure 1181714431 M * daniel_hozac looks like. 1181714447 M * daniel_hozac trippeh: there you go ;) 1181714681 J * FireEgl FireEgl@4.0.0.0.1.0.0.0.c.d.4.8.0.c.5.0.1.0.0.2.ip6.arpa 1181714687 M * slacker403 what kerrnel should i download ? 1181714772 M * doener Bertl: hm, weird, a test connection to irc.oftc.net using netcat died off without seeing a "ping" from the server... 1181714838 M * Bertl yep, they raised the ping interval to 5min 1181714869 M * Bertl while the socket timeout seems to be 5min too 1181714896 M * Bertl but I have adjusted that for pinky, so we just have to wait and see 1181714917 M * slacker403 whihc kerrnel should i DL ? 1181714922 M * slacker403 stable ? 1181714935 M * slacker403 which one has that bind issue fixed ? 1181714985 M * Bertl the patch I linked when you asked, is the stable patch for the kernel you wanted to use :) 1181714997 M * Bertl this is also the one you should download :) 1181715102 M * Bertl the bind issue is an userspace issue in the bind code, so to 'fix' it, you need to rebuild bind (with patches), but all recent kernels contain the kernel side workaround for this issue 1181715105 M * slacker403 does it have that problem fixed i had before ? 1181715116 M * nox Bertl: y? rc1? this patch allready so old? 1181715144 Q * FireEgl Read error: No route to host 1181715150 M * Bertl nox: nah, is the same as the vs2.2.0 patches, just less tested 1181715181 M * Bertl nox: we will probably make it final soon 1181715194 M * nox ic thx btw still very happy with my 2.6.20.4-vs2.2.0 :D! thx 4 the great work and support 1181715221 M * Bertl you're welcome! and thanks for the positive feedback! 1181715259 J * FireEgl FireEgl@4.0.0.0.1.0.0.0.c.d.4.8.0.c.5.0.1.0.0.2.ip6.arpa 1181715379 J * dna ~naucki@130-238-dsl.kielnet.net 1181715433 M * slacker403 Bertl: so is that one problem fixed ? :P 1181715470 M * Bertl which problem? 1181715554 Q * FireEgl Read error: No route to host 1181715897 M * slacker403 bind 1181715899 M * slacker403 capset 1181715899 M * slacker403 thing 1181715908 M * slacker403 is that fixed with that patch ? 1181715959 M * Bertl no, but (as I said) it is worked around 1181716012 M * slacker403 which patch fixes that ? 1181716026 M * daniel_hozac http://people.linux-vserver.org/~dhozac/p/m/bind-9.3.2-caps-when-available.patch 1181716063 M * daniel_hozac but you don't _need_ that with a 2.1+ kernel. 1181716090 J * FireEgl FireEgl@Sebastian.Atlantica.DollarDNS.Net 1181716132 N * DoberMann_ DoberMann[PullA] 1181716147 Q * jannic Ping timeout: 600 seconds 1181716219 M * slacker403 daniel_hozac: whats that patch for ? 1181716221 M * slacker403 the kerrnel ? 1181716223 M * daniel_hozac BIND. 1181716240 A * slacker403 cries 1181716264 M * slacker403 what version fixed it all around ? 1181716428 M * slacker403 i thought you guys fixd it in the kerrnel 1181716483 M * ard off to reboot next server 1181716487 M * ard O/~ 1181716492 Q * ard Quit: rebooting 1181716555 J * jannic ~jan@gasko.hitnet.RWTH-Aachen.DE 1181716822 J * ard ~ard@gw-cistron.kwaak.net 1181716841 M * Bertl slacker403: once again: the issue itself can only be fixed properly in bind, but, as we stated several times, all recent kernels contain a workaround, which doesn't need the fix to bind 1181717044 J * nkukard ~nkukard@dsl-240-121-104.telkomadsl.co.za 1181717173 M * slacker403 Bertl: i want the kerrnel 1181717177 M * slacker403 wit hthe fixaround 1181717191 M * Bertl go for it! 1181717239 J * infowolfe ~infowolfe@c-67-164-195-129.hsd1.ut.comcast.net 1181717254 M * slacker403 Bertl: so will the last stable release one work ? 1181717264 M * Bertl yep, definitely! 1181717366 M * slacker403 how do i apply the patch ? Bertl 1181717367 M * slacker403 :P 1181717403 M * Bertl cd /path/to/kernel && patch -p1 1181722738 Q * Aiken Remote host closed the connection 1181723138 J * dlezcano1 ~dlezcano1@blueice1n1.uk.ibm.com 1181723240 J * shedi ~siggi@ftth-237-144.hive.is 1181723670 J * Aiken ~james@ppp121-45-192-48.lns1.bne1.internode.on.net 1181724569 N * mjt_drunk mjt 1181725975 J * lilalinux ~plasma@dslb-084-058-215-074.pools.arcor-ip.net 1181727041 Q * sannes Remote host closed the connection 1181728579 M * trippeh daniel_hozac: Yay :) 1181729107 J * sannes ace@har.sagt.no 1181729347 J * rgl ~Rui@84.90.10.107 1181729350 M * rgl hello 1181729392 M * rgl can I use vserver host on 64 bits with a vserver guest in 32 bit? 1181729835 M * Guy- yes 1181730060 J * SoftIce ~radio@vc-196-207-45-253.3g.vodacom.co.za 1181730077 M * SoftIce bonjour! i've come across a problem with vserver on feisty 1181730092 M * SoftIce it seems it cant find capabilties being enabled in the kernel 1181730104 M * SoftIce is there a sysctl option or something where I enable capabilties? 1181730223 M * cehteh you need to enable that compile option 1181730239 M * cehteh there is a menu for that (forgot the location, quite at the end afaik) 1181730268 M * SoftIce this same kernel works on edgy 1181730309 Q * mjt Remote host closed the connection 1181730322 J * mjt ~mjt@nat.corpit.ru 1181730365 M * SoftIce isn't there some modprobe capability or something? 1181730418 M * daniel_hozac SoftIce: you can't compile a Linux-VServer kernel without capabilities. 1181730429 M * daniel_hozac and you _really_ don't want your guests modproving things. 1181730446 M * SoftIce http://paste.linux-vserver.org/2558 1181730467 M * SoftIce daniel_hozac: but I thought the vserver patch enables the capabilities 1181730478 M * daniel_hozac that's just because your utils are segfaulting all the time. 1181730486 M * daniel_hozac most likely due to the broken gcc in Ubuntu. 1181730511 M * SoftIce or using an incorrect version? 1181730519 M * daniel_hozac no. 1181730532 M * SoftIce hrm, any sugestions, re-building gcc? 1181730544 M * daniel_hozac IIRC you need to recompile and patch dietlibc, recompile the utils with -fno-stack-protector-all or something like that. 1181730578 M * SoftIce daniel_hozac: so i shouldn't use this package maintenece version of util-vserver anyway? 1181730595 M * SoftIce ii util-vserver 0.30.212-1 user-space tools for Linux-VServer virtual p 1181730596 M * SoftIce root@shaitan:/boot# 1181730697 M * SoftIce daniel_hozac: hrm, dietlibc isn't even installed 1181730718 M * SoftIce let me build it remove this package version and have a look how this utils builds with --fno-stack-protector-all 1181730752 M * SoftIce daniel_hozac: anyone else complained that util-vserver is broken with feisty ? 1181731061 M * SoftIce bwahahha, watching theeeeeee most funniest thing on tv :) feels good to laugh this hard 1181731097 M * SoftIce daniel_hozac: why I ask about this with regards to feisty, as if nobody has complained maybe I should report a bug 1181731801 M * SoftIce **** 'beecrypt' could not be found; 1181731802 M * SoftIce **** this will disable the build of 'vhashify' 1181731810 M * SoftIce should I cancel now and install or can i install after the build? 1181731928 Q * nkukard Quit: Leaving 1181732362 M * daniel_hozac better to install now so you get vhashify built... 1181732370 M * daniel_hozac Ubuntu is known to be broken. 1181732374 M * daniel_hozac nobody has bothered to fix it though 1181732478 M * rgl Guy-, thx. do you have a setup like that? 1181732508 M * Guy- rgl: yes 1181732523 M * rgl Guy-, cool :D 1181732529 M * Guy- quite :) 1181732531 M * SoftIce daniel_hozac: erk, any idea what the configure prefixes look like to match that of ubuntu 1181732553 M * SoftIce as everything have been installed in different locations to the currect configuration i have 1181732630 M * daniel_hozac --sysconfdir=/etc --prefix=/usr --localstatedir=/var i guess. 1181732634 M * daniel_hozac (or something similar) 1181732648 M * SoftIce :) 1181732787 M * SoftIce ok i'm getting closer! 1181732793 M * Guy- rgl: I run Lotus Domino under Debian etch in a 32bit vserver running on a 64bit Debian sid host - I don't think it gets much more unsupported than that :) 1181732815 M * Guy- perhaps if I ran all of this under vmware 1181732836 M * SoftIce daniel_hozac: hopefully last problem! will only know once i resolve it http://paste.linux-vserver.org/2559 1181732843 M * rgl Guy-, heheh 1181733013 M * daniel_hozac SoftIce: /etc/init.d/vprocunhide start 1181733067 M * SoftIce daniel_hozac: could this be a grsec issue? 1181733077 M * SoftIce addtional restrictions to /proc? 1181733077 M * daniel_hozac well, yes. 1181733365 M * SoftIce hrm, what version does this vprocunhide come with? 1181733454 M * SoftIce never mind... 1181733535 M * daniel_hozac all of them... 1181733758 M * SoftIce hrm, what prefix defines Rootdir ? so to set to /var/lib/vservers? 1181733783 M * daniel_hozac --with-vrootdir 1181733796 M * SoftIce ahh, thanks 1181733811 J * yarihm ~yarihm@whitehead2.nine.ch 1181733912 M * SoftIce daniel_hozac who wrote vprocunhide ? 1181733941 M * daniel_hozac Enrico or Jacques, probably. no idea who. 1181734482 M * SoftIce http://paste.linux-vserver.org/2560 1181734489 M * SoftIce daniel_hozac: yes I did follow the instructions to fix it 1181734497 M * SoftIce thing is its enabled in sysctl and i've set the proc info 1181734511 M * SoftIce this is the first time i've had it even though the correct modifictions have been made 1181734524 M * daniel_hozac you are using 0.30.213 and have run the util-vserver initscript, yes? 1181734588 M * SoftIce http://paste.linux-vserver.org/2561 1181734598 M * SoftIce correct for first question, second question output above 1181734621 M * SoftIce can I just ignore those errors? 1181734634 M * daniel_hozac no. 1181734667 M * SoftIce haha 1181734697 M * SoftIce http://paste.linux-vserver.org/2562 1181734729 M * SoftIce ahh, the sym link is incorrect 1181734733 M * daniel_hozac yep. 1181734739 M * daniel_hozac looks like most of your symlinks are incorrect. 1181734779 M * SoftIce i guess i must re-compile with the correct ./configure string 1181734839 M * SoftIce hrm, looks like i lost some of my files maybe with the deinstallation of the util-vserver package in ubuntu 1181734888 M * SoftIce lrwxrwxrwx 1 root root 36 2007-04-25 12:42 vdir -> /etc/vservers/.defaults/vdirbase/nfs root@shaitan:/etc/vservers/.defaults/vdirbase# pwd && ls 1181734889 M * SoftIce root@shaitan:/etc/vservers/.defaults/vdirbase# 1181734917 M * daniel_hozac where does vdirbase point? 1181735033 M * matti Morning. 1181735067 M * SoftIce http://paste.linux-vserver.org/2563 1181735777 M * SoftIce ./configure --help | grep vdirbase 1181735791 M * SoftIce nothing 1181735810 M * Hollow vrootdir 1181735814 M * Hollow :) 1181735819 M * SoftIce oh, same thing 1181735828 M * SoftIce thanks 1181735946 M * SoftIce Hollow does it follow itself, so if say specifiy /etc it will add the extra trailers eg: /etc/vservers? 1181735964 M * SoftIce errr, I was meaning /var/lib && /var/lib/vservers 1181736004 M * Hollow vrootdir is absolute .. 1181736068 M * SoftIce what 1181736068 M * SoftIce all i want is my old vservers working 1181736164 M * SoftIce grrrrrrr 1181736210 M * SoftIce vdir -> /etc/vservers/.defaults/vdirbase/nfs 1181736216 M * SoftIce to get that symlink working Hollow 1181736250 M * SoftIce should I have --with-vrootdir=/etc/vservers 1181736257 M * SoftIce or is this vserver just stuffed and I have to re-build? 1181736278 M * Hollow just change the vdirbase symlink? 1181736287 M * SoftIce beause I don't seem to have the sym links for vdir, cache and run 1181736292 M * SoftIce Hollow: yes, but to where? 1181736301 M * Hollow where you want your vservers to be? 1181736307 M * Hollow this is your choice :) 1181736319 M * SoftIce well it lives in /var/lib/vservers now 1181736342 M * SoftIce so when i ./configure i have it set with ./configure --sysconfdir=/etc --prefix=/usr --localstatedir=/var --with-vrootdir=/var/lib/vservers ? 1181736822 M * SoftIce ok 1181736881 M * SoftIce http://paste.linux-vserver.org/2564 this is what I get once i sym linked my vdir and try to start my vserver 1181736893 M * SoftIce any sugestions, guess i'm getting every problem known to vservers with this :) 1181737143 M * Hollow i guess you need to configure a static context id 1181737151 Q * neuralis Ping timeout: 600 seconds 1181737308 M * SoftIce a static context id? any documentaiton on that ? 1181737363 M * Hollow echo 123 > /etc/vservers//context 1181737368 M * Hollow and --context with vserver-build 1181737382 M * Hollow vserver - build --help 1181737441 M * SoftIce Hollow: ye, i know about it with the build 1181737451 M * SoftIce but i allready have the build 1181737462 M * Hollow echo 123 > /etc/vservers//context 1181737513 M * SoftIce well done 1181737530 M * SoftIce :) 1181737535 M * SoftIce but what has this context got to do with anything? 1181737569 M * SoftIce at dam last! :) thanks everything for having the patience with me! 1181737610 M * Hollow well, the context id the basis of everything 1181737665 M * SoftIce and all it must be is unique ? 1181737682 M * Hollow yep, and between 2-49151 (iirc) 1181737708 M * SoftIce so in theory you can have 49149 vservers? 1181737729 M * daniel_hozac in theory you can have 65533 guests. 1181737756 M * SoftIce oh, so how did i come up with my number? 1181737786 M * SoftIce daniel_hozac: how do you continue if you run out of contexes? 1181737791 M * daniel_hozac you don't. 1181737805 M * daniel_hozac but with the proper settings you can use 2-65535. 1181737809 M * Hollow you run out of resources way before it :) 1181737814 M * SoftIce mwhaha 1181737816 M * SoftIce Hollow really? 1181737817 M * SoftIce :) 1181737821 M * Hollow yeah 1181737825 M * Hollow :) 1181737855 M * SoftIce i'd love to see some resource 1181737863 M * SoftIce benchmarks on what somebody has pushed out of some systems 1181737918 M * SoftIce i've got a 200mhz box with 3 vservers and running things from ftpd, postfix/ mysql / imap/ pop3/ pop3-ssl etc etc and load average is 0.0 1181737940 M * daniel_hozac that all depends on how loaded your services are... 1181737952 M * daniel_hozac obviously you can run a ton of things if none of them are doing anything. 1181737998 M * SoftIce true 1181738483 M * SoftIce weeeeeeeeee i'm so happy i'm finally using this custom kernel with a working vserver 1181739605 Q * Aiken Quit: Leaving 1181740139 Q * sharkjaw Quit: Leaving 1181741832 J * ema ~ema@rtfm.galliera.it 1181742253 Q * dlezcano1 Read error: Connection reset by peer 1181742827 J * orzel ~orzel@freehackers.org 1181742852 M * orzel hello. a new context has just started, that i've never seen : http://rafb.net/p/2Flhrr53.html 1181742876 M * orzel i was wondering, what this is, why it started only 12m ago (i was not even connected to this computer). 1181742899 M * orzel most importantly, should i panic about somebody doing something bad on my computer? 1181743050 J * dlezcano1 ~dlezcano1@blueice2n1.uk.ibm.com 1181743245 N * Bertl_zZ Bertl 1181743249 M * Bertl morning folks! 1181743272 M * Bertl orzel: which one? 1181743278 M * orzel ctx 1 1181743287 M * orzel 'monitoring server' 1181743294 M * orzel that i've never seen, and which is not here anymore now. 1181743295 M * Bertl that is the spectator context 1181743311 M * Bertl i.e. the context used to run things like vserver-stat 1181743329 M * orzel how can it reach 12m of uptime then ? 1181743363 M * Bertl I think that is a bug, as the context does not really exist 1181743375 M * Bertl what version of util-vserver do you use? 1181743376 M * orzel is that something like "when you do vserver-stat, it may, or may not appear" ? like when you do "ps aux | grep ps -w" ? 1181743394 M * orzel i use sys-cluster/util-vserver-0.30.212-r1 (gentoo) 1181743431 M * Bertl yes, that is probably something similar ... I would suggest to upgrade to 0.30.213 (and/or contact/ask) daniel_hozac 1181743597 M * Bertl it could as well be, that vserver-stat gets the uptime right and there was some process 'stuck' in the spectator context for 12 minutes ... 1181743613 M * Bertl (but better double check with the maintainer :) 1181743630 M * orzel i had to reboot my computer because a lot of weird stuff was happening 1181743646 M * orzel like nfsd oopsed, and the corresponding udp/tcp ports were blocked 1181743668 M * orzel i couldn't logout, i couldn't start nfs again. dmesg full of garbage 1181743674 M * Bertl well, that could explain a stuck process ... did you record the oops? 1181743675 M * orzel i suspect the vserver patch to be responsible of that 1181743685 M * orzel they are in my messages i think 1181743700 M * Bertl if so, please upload and we will have a look at it 1181743712 M * Bertl (i.e. we can probably tell you what caused it ...) 1181743723 M * orzel mmh, for some reason i had disabled sys-cluster/util-vserver-0.30.213 explicitely, probably because i've had some problems with this version 1181743818 M * orzel http://rafb.net/p/con81C75.html <- like this ? 1181743860 M * Bertl yep 1181743890 M * Bertl this doesn't look like a Linux-VServer issue at first glance, but I'd advise to upgrade the kernel 1181743912 M * Bertl we had up to rc21 before the final vs2.2.0 was released (a few month ago) 1181743936 M * Bertl (so this version explicitely contains some known Linux-VServer issues :) 1181743982 M * Bertl I would opt for the proprietary nvidia module ... 1181744032 M * Bertl but of course, could be anything else, 00100104 looks like a poisoning pattern 1181744205 M * Bertl orzel: is that the first of those 'BUG' messages you see? 1181744228 M * Bertl (I mean, are there other, similar messages right before that?) 1181744276 M * Bertl for the record, ebx=0x00100100 is LIST_POISON1 ... 1181744380 M * orzel the nvidia module is not used i think. X not even started. 1181744390 M * orzel i did that 3 month ago to watch a divx :) 1181744409 M * orzel Bertl: those were the first ones, ys 1181744410 M * orzel yes 1181744413 M * orzel what is poisoning ? 1181744424 M * orzel some pattern meant to be recognised ? 1181744444 M * Bertl the kernel has mechanisms to maintain lists and other common objects 1181744480 M * Bertl the debug mechanisms 'poison' removed or replaced pointers in those elements with a recognizeable value 1181744497 M * Bertl (which triggers an oops, when used/dereferenced) 1181744522 M * Bertl the value 00100104 is 0x00100100 + 4 1181744545 M * Bertl i.e. it was likely to be cause by dereferencing such a poisoned pointer 1181744560 M * orzel linux-2.6.20-vserver-2.2.0 <- i'm gonna update to that. seems ok, no ? 1181744582 M * Bertl seems okay so far, yes 1181744585 M * orzel Bertl: ok, i understand the poison thing. didn't know about this. 1181744659 M * orzel i dont need Kernel-based Virtual Machine (KVM) support, nor the hypevision stuff, right ? It's getting confused with all those different (para)virtualisation stuff 1181744693 M * orzel ah, VSERVER_WARN is cool :) 1181744738 M * Bertl nope, you don't need kvm for Linux-VServer ... 1181744766 M * orzel :) 1181744768 M * orzel thx 1181745415 Q * lilalinux Remote host closed the connection 1181746088 Q * rob-84x^ Ping timeout: 600 seconds 1181746651 Q * huiz Ping timeout: 600 seconds 1181747310 Q * orzel Read error: Connection reset by peer 1181747340 J * sourcerer ~philipp@email.falcontrol.com 1181747343 M * sourcerer Hi! 1181747393 M * sourcerer Does anyone have an example config how I can have a host with several network cards, and I can bind the vserver guests to specific network cards, and do the routing properly? 1181747444 M * Bertl welcome sourcerer! 1181747452 M * sourcerer Hallo Bertl! 1181747458 M * Bertl first, there is no _binding_ to network cards in Linux-VServer ... 1181747484 M * sourcerer Why do I have to choose a network card for each guest, and the guest only sees that network card then? 1181747488 M * Bertl Linux-VServer uses IP isolation (layer 3) so you can only bind your guests to those 1181747513 M * Bertl if and only if you choose a network card, the tools will add/remove the ip for you 1181747531 M * Bertl if you do not specify a nic, the tools assume that you are maintaining the ip yourself 1181747581 M * Bertl the guests only see nics which carry IPs assigned to them (to make it simpler and more natural for the guest) 1181747698 M * sourcerer Well, I chose the network card and the IP myself, which works fine. I only have the problem that the guests don´t get/have a default-route, and I haven´t found a way to assign them one. 1181747726 M * sourcerer (Which isn´t a problem for simple webservers and similar things, but IRC-servers and mailserver prefer to have internet access) 1181747771 M * Wonka hmm, my vserver has a default route... 1181747777 M * Bertl sourcerer: there is a difference between having separate routing tables (which can easily be done with advanced routing setups) and 'providing internet access' 1181747822 M * Bertl sourcerer: if you really have more than one 'default' gateway (i.e. different defaults for host and guest) you have to use multiple routing tables 1181747842 M * sourcerer I read that Debian Etch has problems with the advanced routing stuff. Do you know, whether this kind of setup is affected? 1181747844 J * rob-84x^ rob@submarine.ath.cx 1181747880 M * Bertl sourcerer: no idea about debian, what issues does it have? 1181747934 M * sourcerer http://mailman.ds9a.nl/pipermail/lartc/2006q4/019898.html (I am not sure at the moment, whether that´s the right posting) 1181747977 M * sourcerer Yes, I need different defaults for host and guest 1181748006 M * Bertl doesn't seem related (the issue) but you should definitely use a newer kernel anyway :) 1181748013 J * hallyn ~xa@adsl-75-0-158-74.dsl.chcgil.sbcglobal.net 1181748069 M * nox hey sourcerer :D 1181748100 J * stefani ~stefani@tsipoor.banerian.org 1181748106 M * Bertl morning stefani! 1181748117 M * stefani morgen. 1181748122 M * sourcerer Is it possible to get the advanced routing integrated into the vserver scripts, so that vserver automatically creates the multiple routing tables on demand? 1181748149 M * Bertl you probably can do that, if you assign a table number per guest 1181748229 J * pmenier ~pme@LNeuilly-152-22-72-5.w193-251.abo.wanadoo.fr 1181748246 M * pmenier Hi 1181748265 M * Bertl wb pmenier! 1181748277 Q * rgl Ping timeout: 600 seconds 1181748332 M * nox sourcerer: where do you see the security advance having different gw? or other reason? 1181748349 M * Bertl probably he has more than one gateway ... 1181748364 M * sourcerer The network cards are in different networks 1181748442 M * nox probably snat is enough then? 1181748529 M * nox or if you have a public ip for every guest it should work ootb 1181748689 M * nox which kernel/utils? 1181748846 J * neuralis ~krstic@solarsail.hcs.harvard.edu 1181748922 M * Bertl wb neuralis! 1181748930 M * sourcerer nox: Linux 2.6, Debian Etch, standard vserver 1181748932 M * Bertl sourcerer: http://archives.linux-vserver.org/200311/0470.html 1181748932 M * neuralis 'morning 1181750143 Q * dlezcano1 Read error: Connection reset by peer 1181750235 J * rgl ~Rui@84.90.10.107 1181750406 N * ensc Guest1804 1181750416 J * ensc ~irc-ensc@p54B4E5C6.dip.t-dialin.net 1181750647 Q * Guest1804 Ping timeout: 600 seconds 1181750974 M * pmenier Bertl : do you think i need to apply delta-pid_task-fix01.diff to patch-2.6.21.5-vs2.2.0-rc1.diff 1181751001 M * pmenier cf hell's bug :-) 1181751037 M * Bertl nope, it is already included 1181751056 M * pmenier ok thanks 1181751384 M * SoftIce hello, does anyone have any idea how i can find out what is causing a system to seg fault on boot, i get around 30 seg faults then it just carries on as usual 1181751440 M * SoftIce right after it loads initrd.img i believe 1181751440 J * bonbons ~bonbons@2001:5c0:85e2:0:20b:5dff:fec7:6b33 1181751446 M * SoftIce as its right after it unpacks the image 1181751467 M * Bertl SoftIce: 64bit with 32bit hotplug? 1181751491 M * SoftIce 32bit all the way 1181751512 M * Bertl including the kernel? what segfaults? 1181751523 M * SoftIce Bertl I don't know ? 1181751531 M * SoftIce thats what i wanted to know how to determine 1181751542 M * Bertl well, no kernel info? (dmesg) 1181751598 M * SoftIce Bertl i'll paste my dmesg info 1181751611 M * Bertl (please use paste.linux-vserver.org for everything longer than 3 lines) 1181751650 M * Bertl daniel_hozac, doener: ping? 1181751821 M * doener pongo 1181751881 M * Bertl daniel_hozac, doener: I have a number of whitespace cleanups to vs2.2.0, and I would appreciate verification ... 1181751895 M * SoftIce http://paste.linux-vserver.org/2570 1181751924 M * Bertl daniel_hozac, doener: the main purpose of those cleanups is to make the code more conformant to Linux kernel coding style 1181751946 M * doener URL? 1181751954 M * Bertl will upload it shortly 1181752004 M * Bertl SoftIce: last line looks like grsec denying something .. maybe grsec blocks certain stuff? 1181752048 M * SoftIce Bertl: ye that was me just denying mode +s in vserver 1181752053 M * SoftIce i was going to remove that restriction 1181752072 M * SoftIce got alot of device mapper issues 1181752113 Q * rgl Ping timeout: 600 seconds 1181752145 M * SoftIce Bertl: its failing so bad that i'm getting this 1181752159 M * SoftIce . /dev/hdc1 on / type ext3 (rw,errors=remount-ro) 1181752174 M * SoftIce or am i reading that incorrectly. has it re-mounted my device ro ? 1181752193 M * Bertl no, looks fine, just says: if error, then remount ro 1181752198 M * SoftIce ahh 1181752220 M * slacker403 im back 1181752229 M * Bertl wb slacker403! 1181752243 M * SoftIce Bertl: is it safe to disable lvm and evms ? 1181752260 M * SoftIce or does initrams relay on lvm 1181752264 M * slacker403 Bertl: back to compiling kerrnel ..........i might get it done today 1181752269 M * Bertl SoftIce: not if you have your boot system on one of them :) 1181752312 M * SoftIce Bertl you ever experience this device-mapper: table: 253:0: linear: dm-linear: Device lookup failed ? 1181752314 J * meandtheshel1 ~markus@85-124-39-174.dynamic.xdsl-line.inode.at 1181752329 M * SoftIce i'm sure its directly related to udev? 1181752501 J * rgl ~Rui@84.90.10.107 1181752523 M * Bertl SoftIce: no, haven't seen those .. but could be udev related 1181752566 M * Bertl daniel_hozac, doener: http://vserver.13thfloor.at/Experimental/delta-mainline-clean01.diff 1181752985 M * SoftIce http://paste.linux-vserver.org/2571 1181752991 M * SoftIce any idea why my drive is running in 16bit mode? 1181753018 M * Bertl how is that Linux-VServer related? :) 1181753050 M * SoftIce well because in my vserver i'm getting alot of errors with my zaptel drivers 1181753064 M * Bertl SoftIce: anyway, try hdparm -u 1 -c 1 -m 16 /dev/hdc 1181753128 M * Bertl but that is just performance tuning, it will not change anything for the guests (except for performance) 1181753297 M * SoftIce ye, i'm getting buffer over run errors. 1181753317 M * SoftIce hopefully this helps then 1181753327 M * Bertl well, then your disk/IO subsystem is probably boken 1181753349 M * SoftIce Bertl could it be due to no support for the controller? 1181753367 M * SoftIce this happens on a few of my vservers, yet some of them work 1181753372 M * SoftIce and its using different hardware 1181753384 M * Bertl could be ... not sure what mainline issue you are tripping over 1181753472 M * SoftIce similar to this error 1181753474 M * SoftIce Mar 8 18:01:00 ciop kernel: zaphfc: bchan rx fifo not enough bytes to 1181753474 M * SoftIce receive! (z1=528, z2=527, wanted 8 got 2), probably a buffer overrun 1181753501 M * Bertl sounds more like your network card 1181753510 M * pmenier Bertl: just for info : 2.6.21.5-rc1 works fine for me on amd64. I'll will try it on another box friday. 1181753522 M * Bertl pmenier: excellent! tx! 1181753526 M * SoftIce Bertl i've even disabled the NIC, ahh you mean the isdn modem ? 1181753557 Q * pmenier Quit: Quitte 1181753564 M * Bertl whatever zaphfc is :) 1181753564 M * SoftIce ye could be, shouldn't really try get away with a 150bux modem instead of a proper 5000rand bri card 1181753601 M * SoftIce thanks 1181753607 M * Bertl np 1181754203 M * doener Bertl: in #define init_vx_info, shouldn't p and i be in parentheses? 1181754254 M * doener same for set_vx_info 1181754255 M * Bertl I thought about that, but the question is, why? 1181754304 M * Bertl those are complete macro arguments passed to the next function/inline ... I don't see how they could break? 1181754541 M * doener true, comma expressions should appear in parentheses anyway 1181754689 Q * SoftIce Read error: Connection reset by peer 1181754892 M * Bertl (otherwise they will be passed as separate arguments in the first place :) 1181755099 M * doener Bertl: is it just the leading + that makes vx_sub_cres appear to have one tab too much? 1181755146 M * doener yeah, it is 1181755163 Q * FireEgl Read error: No route to host 1181755174 M * doener reviewing in a browser window is a stupid idea :) 1181755640 Q * ema Quit: leaving 1181756671 Q * yarihm Quit: This computer has gone to sleep 1181756879 M * doener Bertl: hm, there are a few instances of &(m->v), does that make sense? I'm used to seeing &(m)->something, so I'd expect that to be &(m)->(v) instead. 1181756883 M * Hollow fyi, very interesting article: http://portal.acm.org/citation.cfm?id=1030124&dl=ACM&coll=&CFID=15151515&CFTOKEN=6184618 1181756907 M * Bertl doener: could you point me to that, please? 1181756950 M * doener vx_acc_* 1181756986 M * doener hm, vx_acc_page* actually, e.g. vx_acc_pages 1181757037 M * Bertl guess we want both there, i.e. &((m)->v) 1181757078 M * Bertl OTOH, & as unary operator should be fine without () 1181757089 M * Bertl dilinger: could you give us a hint here? 1181757101 M * doener what about v? no parentheses required there? 1181757138 M * Bertl well, that is replaced by the corresponding substruct, so that cannot be messed with ... 1181757151 M * doener k 1181757155 M * dilinger Bertl: hm? 1181757167 M * dilinger oh, precedence? 1181757182 M * Bertl well, the question is, do we need/should we use: 1181757193 M * Bertl &((m)->v) or &(m)->v 1181757200 M * dilinger &(m)->v is fine 1181757212 M * Bertl okay, tx! 1181757747 M * doener up to line 903, everything else looks fine. Will continue later, somehow I get a bit dizzy today by looking too closely at such small changes :-/ 1181757760 M * Bertl okay, np, tx! 1181757790 M * Bertl dilinger: is ', void __user *data)' or ', void __user * data)' correct? 1181757843 M * dilinger Bertl: i'm not sure; i see lots of both. the important thing is that the style is consistent 1181757851 M * dilinger though personally, i prefer the former 1181757876 M * Bertl okay, me too, just Lindent (script) does this one different 1181757898 M * Bertl (but as it seems only for such cases as __user) 1181757912 M * dilinger i bet lindent is getting confused 1181757924 M * Bertl could be, anyway, we stick with the former 1181757954 Q * Baby Quit: byes! 1181758007 M * dilinger Bertl: i hear you're working on a new release? 1181758045 M * Bertl yep, we will soon have a special olpc branch and an updated vs2.{2,3}.x 1181758095 J * coderanger_ ~laptop@wireless-19-27.media.mit.edu 1181758101 N * coderanger_ coderanger 1181758128 M * coderanger daniel_hozac: Ping 1181758362 Q * shedi Quit: Leaving 1181758949 M * Bertl okay, off for now .. back a little later ... 1181758953 N * Bertl Bertl_oO 1181759924 J * phedny_ ~mark@ip56538143.direct-adsl.nl 1181759960 N * DoberMann[PullA] DoberMann 1181759962 M * daniel_hozac coderanger: pong. 1181759975 M * daniel_hozac coderanger: btw, just FYI, you were looking at the wrong libvserver yesterday... 1181759983 M * coderanger daniel_hozac: hrm? 1181759994 M * coderanger Which is the right onw 1181759996 M * coderanger er, one 1181760010 M * daniel_hozac http://svn.linux-vserver.org/projects/libvserver is not the one in util-vserver. 1181760035 M * daniel_hozac the one in util-vserver is at http://svn.linux-vserver.org/projects/util-vserver/browser/trunk/lib/vserver.h 1181760059 J * Ashsong ~mstone@pepper.cs.swarthmore.edu 1181760082 M * Ashsong daniel_hozac: Hi. I'm also working with coderanger. 1181760090 M * daniel_hozac right, i saw that. 1181760101 M * Ashsong daniel_hozac: You said we're using an incorrect version of libvserver? 1181760122 M * daniel_hozac at least if you want to wrap the one in util-vservber. 1181760124 M * coderanger daniel_hozac: Is there a reason to use one vs. the other? 1181760126 M * daniel_hozac +spelling 1181760141 M * daniel_hozac the former is Hollow's syscall wrappers, pretty much. 1181760155 M * daniel_hozac which lack the compatibility stuff. 1181760179 M * Ashsong Does "the former" refer to SVN trunk libvserver? 1181760193 M * daniel_hozac while the one in util-vserver, well, comes with util-vserver and has all of the compat stuff. 1181760198 M * daniel_hozac Ashsong: yes. 1181760219 M * daniel_hozac as i said just before you joined, http://svn.linux-vserver.org/projects/util-vserver/browser/trunk/lib/vserver.h is what you want... 1181760228 M * Ashsong daniel_hozac: thanks. 1181760286 M * Hollow coderanger: well, it doesn't make much difference... since all syscalls may return -1, and xid_t is unsigned, it's a bit more intuitive from my pov ... 1181760358 M * coderanger Hollow: Except that int may not be 32 bit on all platforms 1181760366 M * Hollow so? 1181760374 M * Hollow why should that matter here? 1181760386 M * coderanger If you have >65k vservers, that could break 1181760398 M * Hollow context ids only range from 0 to 65536 1181760409 M * coderanger Then why are they a uint32? :P 1181760415 M * Hollow no idea 1181760445 Q * phedny Ping timeout: 600 seconds 1181760472 M * coderanger in any case, if you mean for it to be a uint16, you should make it that 1181760482 M * daniel_hozac it should be a uint32. 1181760486 M * Hollow i do what the kernel does 1181760490 M * Hollow :) 1181760604 N * phedny_ phedny 1181760845 M * coderanger daniel_hozac: If we are going through libvserver for everything, do we actually need the rest of the utils? 1181760870 M * daniel_hozac coderanger: i guess not. 1181761108 M * Hollow coderanger: i appreciate that :) 1181761488 J * tanjix ~tanjix@office.star-hosting.de 1181761528 M * coderanger daniel_hozac: Are there any API differences (or anything else) that would make it a good idea to use util-versver from trunk 1181761564 M * Hollow the api is completely incomaptible .. 1181761574 M * daniel_hozac yeah. 1181761583 M * daniel_hozac function names differ, the arguments aren't the same, etc. 1181761606 M * coderanger Okay 1181761613 M * daniel_hozac or do you mean compared to util-vserver 0.30.213? 1181761616 M * Hollow i still have the dream that util-vserver will use (my) libvserver one day :P 1181761622 M * coderanger daniel_hozac: Yes 1181761634 M * daniel_hozac Hollow: i still consider util-vserver's libvserver the more complete one... 1181761645 M * Hollow well, depends on the pov :) 1181761648 M * coderanger daniel_hozac: Or .212, which is what is currently packaged for FC7 1181761666 M * daniel_hozac F7 should get 0.30.213 as soon as ensc makes the bodhi request... 1181761696 M * daniel_hozac IIRC there were a few changes in 0.30.213. 1181761761 M * daniel_hozac http://svn.linux-vserver.org/projects/util-vserver/changeset?new=trunk%2Flib%2Fvserver.h%402501&old=trunk%2Flib%2Fvserver.h%402415 1181761815 M * Hollow daniel_hozac: btw, what do you think is missing? 1181761823 M * daniel_hozac from libvserver? 1181761827 M * Hollow yep 1181761828 M * daniel_hozac all the compat stuff :) 1181761833 M * Hollow haha 1181761834 M * Hollow :) 1181762210 P * click [IRSSI] 1181762369 M * coderanger daniel_hozac: Okay, so I guess we should be wrapping trunk 1181762373 P * sourcerer Konversation terminated! 1181762393 M * daniel_hozac well, there have been no changes since 0.30.213. 1181762450 M * coderanger daniel_hozac: Oh, so that diff was back to 212 1181762454 M * daniel_hozac yeah. 1181762486 J * nkukard ~nkukard@dsl-240-121-104.telkomadsl.co.za 1181762530 M * Ashsong daniel_hozac: In order to make this wrapper, I need to know how wide xid_t is actually going to be. 1181762535 M * coderanger daniel_hozac: Okay, so we can do 213, and just make sure that fc7 gets bumped 1181762536 M * Ashsong How do you suggest that I find that out? 1181762559 M * daniel_hozac IIRC unsigned int is 32-bit on Linux. 1181762587 M * Ashsong daniel_hozac: It's merely defined to be (for the sake of Doxygen) an_unsigned_int_t 1181762597 M * Ashsong I'm not sure where the actual definition is. 1181762599 M * daniel_hozac compat.h 1181762606 M * Ashsong thanks. 1181762613 M * daniel_hozac well, that actually sets it to uint32_t. 1181762630 J * click click@ti511110a080-0623.bb.online.no 1181762651 M * Ashsong great. 1181763386 M * coderanger daniel_hozac: When one of the libvserver calls fails, does it stick an error code/message somewhere (errno?)? 1181763429 M * daniel_hozac errno, of course. 1181763448 M * daniel_hozac most of them are just syscall wrappers, with the version wrappers in front. 1181763530 M * coderanger Hrm, I don't suppose you know a good way to get at that from ctypes ;-) 1181763554 M * daniel_hozac errno? not really... 1181763576 M * coderanger I think I found some code for it 1181763726 M * daniel_hozac i guess that depends on which libc you're using. 1181763764 M * coderanger libc.__errno_location().contents.value is what I found 1181763782 M * coderanger but that is returning 0, even though vs.vc_get_version() == -1 1181763923 M * daniel_hozac tried running vserver-info - APIVER? 1181763949 M * daniel_hozac i.e. made sure it works at all? 1181764014 M * coderanger Hmm, segfault 1181764017 M * coderanger lovely 1181764200 M * coderanger Which rev corresponds to .213, it doesn't seem to have a tag? 1181764229 M * daniel_hozac wow, damn. 1181764252 M * daniel_hozac 2541 1181764274 M * coderanger thanky 1181764427 J * mellen ~mellen@0203ds1-sol.0.fullrate.dk 1181764942 M * coderanger daniel_hozac: http://paste.lisp.org/display/42723 1181765002 M * daniel_hozac and, you're running it as root? 1181765015 M * daniel_hozac on a Linux-VServer patched kernel? 1181765027 M * coderanger Not as root 1181765041 M * coderanger Ahh, sudo fixed it 1181765051 M * coderanger Didn't think the version number would require root 1181765067 M * daniel_hozac the entire syscall requires CAP_CONTEXT. 1181765080 M * coderanger makes sense 1181765107 M * coderanger vserver-info still segfaults, even under sudo 1181765124 M * daniel_hozac did you compile the utils yourself? 1181765132 M * coderanger Yes 1181765147 M * coderanger This is .213 on Ubuntu Feisty 1181765156 M * daniel_hozac ah, Ubuntu. 1181765160 M * daniel_hozac that explains it ;) 1181765191 M * coderanger Does this mean I have a yet another reason to say we need a FC dev machine? 1181765211 A * coderanger does not drink the Ubuntu kool aid 1181765214 M * daniel_hozac as i recall, you need to at least compile the utils with -fno-stack-protector-all. 1181765233 M * daniel_hozac possibly recompiling dietlibc with the SSP patch... 1181765261 M * coderanger Will this effect the library? if not I have a strong urge to just ignore it for now 1181765281 M * daniel_hozac most likely. 1181765301 M * coderanger fun, I'll go have fun with the makefiles then 1181765312 M * daniel_hozac but glibc should have the necessary functions... not sure though. 1181765336 M * coderanger I am using the dietlibc from apt 1181765426 M * daniel_hozac IIRC that's broken. but it's been a while since we debugged Ubuntu the last time... it really needs a proper maintainer. 1181765460 M * coderanger What is your primary dev platform. If it is F(6|7) I would just as soon move our dev to that 1181765473 M * coderanger since F7 is what our actual machines run 1181765536 M * daniel_hozac my primary machine is FC5 right now, but my laptop is F7. 1181765560 M * coderanger Okay 1181765570 M * coderanger Thanks for putting up with all my questions :) 1181765590 M * daniel_hozac np, it's what we're here for... ;) 1181765857 J * FireEgl FireEgl@4.0.0.0.1.0.0.0.c.d.4.8.0.c.5.0.1.0.0.2.ip6.arpa 1181766649 J * fatgoose ~samuel@76-10-152-94.dsl.teksavvy.com 1181766649 Q * fatgoose_ Read error: Connection reset by peer 1181766942 J * ema ~ema@rtfm.galliera.it 1181767834 J * phedny_ ~mark@ip56538143.direct-adsl.nl 1181767869 Q * pusling Ping timeout: 480 seconds 1181767994 J * pusling ~pusling@217.74.211.114 1181768043 M * pusling hi. I ifup/downed my main interface. it took all the 'virtual' devices from the vservers with it. how do I up these interfaces again? 1181768060 Q * dna Quit: Verlassend 1181768095 M * daniel_hozac pusling: ip addr add / dev 1181768235 Q * phedny Ping timeout: 480 seconds 1181768240 M * pusling so for a vserver iface for a vserver called 'web' on ip 192.168.1.5, it should be ip addr add 192.168.1.5/web dev eth0 ? 1181768297 M * daniel_hozac no, the prefix is the CIDR prefix... 1181768312 M * daniel_hozac e.g. /24 for 255.255.255.0. 1181768318 M * pusling as in /24 ? 1181768321 Q * bonbons Quit: Leaving 1181768357 J * pusling_ pusling@88.212.70.38 1181768426 M * pusling ah. it worked. pusling_ is here. 1181768433 M * pusling daniel_hozac: thanks. 1181768437 Q * pusling Quit: Konversation terminated! 1181768493 N * pusling_ pusling 1181769047 J * yarihm ~yarihm@84-75-109-39.dclient.hispeed.ch 1181769185 Q * weasel Ping timeout: 600 seconds 1181769323 J * weasel weasel@asteria.debian.or.at 1181769778 Q * mellen Remote host closed the connection 1181769999 J * sauron ~zeus16384@12.20.4.100 1181770003 M * sauron hi guys 1181770018 M * sauron anyone have a clue what spawn.dat is? 1181770102 M * Guy- my first guess is: it's a filename? 1181770109 M * sauron Guy ;) 1181770147 M * sauron maybe... I have an application that is supposed to start another in some case. now inside vserver looks it is not starting it (maybe a fork?) 1181770152 M * sauron but outside vserver it does 1181770158 M * Guy- so strace it 1181770158 M * sauron and when i used strace on the command, 1181770164 M * sauron I see spawn.dat in the output 1181770170 M * sauron but there is no such file anywhere 1181770192 M * Guy- and in what context do you see spawn.dat? I'm damn sure there is no syscall by this name 1181770199 Q * slacker403 Ping timeout: 480 seconds 1181770221 M * sauron sorry, how do I see that context? 1181770239 M * sauron open("spawn.dat", O_RDONLY|O_LARGEFILE) = 14 1181770252 M * sauron open("spawn.dat", O_RDONLY|O_LARGEFILE) = -1 ENOENT (No such file or directory) 1181770258 M * sauron and things like that 1181770286 M * Guy- the first open() succeeds 1181770289 M * daniel_hozac so? 1181770295 Q * duckx Quit: Client exiting 1181770325 M * Guy- sauron: my advice is, compare the strace logs of the case where it works and the case where it doesn't, and find syscalls that failed in the latter case but didn't in the former 1181770338 M * Guy- sauron: then read the man page of that syscall to find out what it was supposed to do 1181770357 M * Guy- sauron: and fix the problem (e.g. if it was trying to open a file that wasn't there, put the file there) 1181770365 M * Guy- it's really not hard, all you need is patience 1181770386 M * Guy- daniel_hozac: Bertl mentioned there might be a patch for the xfs file corruption issue on hashified files? 1181770405 M * daniel_hozac not yet, AFAIK. 1181770424 M * Guy- daniel_hozac: what's causing the problem, btw? 1181770452 M * daniel_hozac XFS isn't using the usual sendfile implementation. 1181770477 M * Guy- uh, maybe we're talking about two different bugs? mine isn't related to sendfile() 1181770503 M * Guy- the problem is if I open a hashified file in joe(1), it suddenly only contains binary zeroes 1181770513 M * Guy- this doesn't happen with vim 1181770529 M * daniel_hozac that is the same problem. 1181770538 M * Guy- how is this related to sendfile? 1181770552 Q * Johnnie resistance.oftc.net oxygen.oftc.net 1181770553 J * Johnnie ~jdlewis@c-67-163-247-109.hsd1.pa.comcast.net 1181770553 M * sauron got you, but thre is no such file 1181770553 M * sauron must be something else 1181770553 M * sauron as in another box the thing works, and I never heard of such file there 1181770553 M * sauron what I suspect... maybe checking of ps output is done in a certain way in vserver that this process doesnt do 1181770553 M * sauron or like 1181770553 M * sauron but I may be wrong 1181770553 M * sauron because it is supposed to check if a process is runing, and start it if not. 1181770570 M * daniel_hozac COW-link breaking uses sendfile. 1181770581 M * Guy- ah 1181770647 M * Guy- sauron: you just pasted a line from an strace log that said the spawn.dat file was there 1181770662 M * Guy- sauron: the process was able to open it on file descriptor 14, in read-only mode 1181770716 M * Guy- sauron: I'm not saying the problem is that the file doesn't exist in the other installation - maybe it's created by the program itself and removed later 1181770794 M * sauron hm 1181770943 M * sauron anyway I sniff further, thanks. looks complex issue to me. 1181771144 J * Aiken ~james@ppp121-45-192-48.lns1.bne1.internode.on.net 1181771819 M * fatgoose hey 1181771826 M * fatgoose vnamespace: vc_xidopt2xid("/etc/vservers/trou"): No such file or directory 1181771836 M * fatgoose after a warm reboot 1181771838 M * sid3windr well, is there? :) 1181771863 M * fatgoose 4 drwxr-xr-x 6 root root 4096 Dec 18 09:13 /etc/vservers/trou 1181771890 M * fatgoose and the v is running 1181771891 M * fatgoose lust:~# vserver-stat | grep trou 1181771891 M * fatgoose 111 16 370.8M 46.3M 0m01s24 0m00s28 1h36m42 trou 1181771958 M * fatgoose oh man, just a filesystem mess... i've got it 1181771960 M * fatgoose sorry 1181772014 Q * rgl Ping timeout: 480 seconds 1181772374 N * DoberMann DoberMann[ZZZzzz] 1181772681 P * marcfiu 1181772788 J * zLinux ~zLinux@88.213.35.226 1181773759 Q * zLinux Remote host closed the connection 1181773886 Q * Ashsong Quit: leaving 1181774362 M * sauron later 1181774370 Q * sauron 1181774739 J * rgl ~Rui@84.90.10.107 1181774813 J * zLinux ~zLinux@88.213.35.226 1181775244 Q * yarihm Quit: Leaving 1181775284 M * coderanger daniel_hozac: Are there some symbols that you aren't including in the library? 1181775293 M * coderanger daniel_hozac: Specificially: vc_new_s_context 1181775307 M * daniel_hozac that depends on how you compiled it. 1181775335 M * daniel_hozac it's a legacy syscall, so it's only available if you built it with --enable-apis=NOLEGACY or --enable-apis=ALL. 1181775379 M * coderanger Roger, I'll rebuild with ALL 1181775386 M * daniel_hozac good luck ;) 1181775411 M * daniel_hozac i'd suggest you go for NOLEGACY. should get you all the possible symbols. 1181775429 M * daniel_hozac (i have not been able to compile ALL for quite some time) 1181775435 M * coderanger Ahh 1181775446 M * coderanger Is there just a list of what is legacy? 1181775455 M * coderanger Since we needn't wrap the legacy functions 1181775484 M * daniel_hozac vc_new_s_context and vc_set_ipv4root. 1181775497 M * coderanger Excellent, we'll just ignore those 1181775499 M * coderanger Thanky 1181776085 Q * ema Quit: leaving 1181776434 Q * zLinux Remote host closed the connection 1181776438 J * fatgoose_ ~samuel@76-10-152-94.dsl.teksavvy.com 1181776442 Q * fatgoose_ 1181776498 J * zLinux ~zLinux@88.213.35.226 1181776851 Q * fatgoose Ping timeout: 480 seconds 1181777599 Q * zLinux Ping timeout: 480 seconds 1181777803 M * mstrobert goodnight 1181777969 Q * ktwilight_ Ping timeout: 480 seconds 1181778925 P * stefani I'm Parting (the water)