1279153662 Q * derjohn_foo Ping timeout: 480 seconds 1279154225 J * derjohn_foo ~aj@d065052.adsl.hansenet.de 1279156025 J * FireEgl FireEgl@Sebastian.FireEgl.Com 1279157129 P * joat Leaving 1279158283 Q * imcsk8 Quit: This computer has gone to sleep 1279162551 J * balbir_ ~balbir@wsp05977122wss.cr.net.cable.rogers.com 1279163377 M * Bertl_oO off to bed now ... have a good one everyone! 1279163382 N * Bertl_oO Bertl_zZ 1279167129 J * petzsch ~markus@dslb-092-078-157-225.pools.arcor-ip.net 1279168067 J * cuba33ci ~cuba33ci@111-240-232-237.dynamic.hinet.net 1279168389 Q * petzsch Quit: Leaving. 1279169406 Q * cuba33ci Remote host closed the connection 1279169407 J * petzsch ~markus@dslb-092-078-157-225.pools.arcor-ip.net 1279169479 J * cuba33ci ~cuba33ci@111-240-216-141.dynamic.hinet.net 1279171017 M * isodude daniel_hozac: It's some of the things I added that didn't work. But I'll fix that now! 1279171638 M * isodude Why is chrootsed() used in debian initpost? Couldn't you aswell use sed -i -e 'expr' ? 1279172018 Q * FireEgl Remote host closed the connection 1279172240 J * imcsk8 ~ichavero@evdomip-67-179.iusacell.net 1279173683 J * ntrs ~ntrs@77.28.160.226 1279174029 J * ntrs_ ~ntrs@77.28.4.41 1279174470 Q * ntrs Ping timeout: 480 seconds 1279174703 J * yarihm ~yarihm@217-162-96-248.dclient.hispeed.ch 1279175086 Q * derjohn_foo Ping timeout: 480 seconds 1279175637 J * dna ~dna@76-198-103-86.dynamic.dsl.tng.de 1279175710 J * FireEgl ~FireEgl@2001:470:e056:1:19a5:6588:697d:fb 1279175872 Q * yarihm Quit: Leaving 1279176695 J * derjohn_foo ~aj@213.238.45.2 1279178495 J * thierryp ~thierry@zankai.inria.fr 1279179157 M * isodude I am wondering about all this failure with locales, when apt-get tries to set locale it fails. I've solved it by doing apt-get install language-pack-en-base, but it should be possible to solve it by doing /usr/sbin/locale-gen (as in the debian initpost-script) this works until like the second apt-get command. Don't understand why it stops working though. 1279179499 N * yeeeepssen AlexBurnsRED 1279179571 Q * AlexBurnsRED 1279180150 M * isodude Ok, I'm posting this last changes I made to the initpost-file. If someone would like to take a look at it please do. Currently the sed -i -e is failing for console output, why it does that I don't know. It's almost complete anyhow. 1279180241 Q * imcsk8 Quit: This computer has gone to sleep 1279185162 J * gnuk ~F404ror@pla93-3-82-240-11-251.fbx.proxad.net 1279188536 Q * dna Quit: Verlassend 1279189581 J * emcepe ~mcp@wolk-project.de 1279189705 Q * balbir_ Remote host closed the connection 1279190041 Q * mcp Ping timeout: 480 seconds 1279190041 N * emcepe mcp 1279191201 Q * ntrs_ Read error: Connection reset by peer 1279191212 Q * petzsch Quit: Leaving. 1279191212 J * ntrs_ ~ntrs@77.28.29.199 1279191352 J * petzsch ~markus@dslb-092-078-157-225.pools.arcor-ip.net 1279193470 J * BenG ~bengreen@cpc6-aztw22-2-0-cust100.aztw.cable.virginmedia.com 1279195158 J * thierryp_ ~thierry@vis175d.sophia.inria.fr 1279195163 Q * thierryp_ Remote host closed the connection 1279195225 Q * thierryp Ping timeout: 480 seconds 1279195336 M * ard daniel_hozac : how do you want to have the debian/ directory? It's not formally finished but I am going to start using it anyway 1279197685 J * Pazzo ~ugelt@reserved-225136.rol.raiffeisen.net 1279199215 J * SauLus ~SauLus@d073133.adsl.hansenet.de 1279199423 N * Bertl_zZ Bertl 1279200047 M * Pazzo Hi Bertl! 1279200055 M * Pazzo Interested in some IPv6-related feedback? 1279200091 Q * BenG Quit: I Leave 1279200321 M * Pazzo a) IPv6 works, I'm running 2.6.32.10-vs2.3.0.36.29.2 in Italy and prepared 2.6.32.16-vs2.3.0.36.29.4 for some first (small) fully productional IPv6-vservers in Germany 1279200404 M * Pazzo b) mask seems to be ignored, I tried setting 48 and 64 in 0/mask - always got /128. 1279200479 M * Pazzo That's not an issue for me as of how my routing is done in Italy (custom tunnel finishes on the same host and is routing the whole IPv6-range directly to it). This will be different in Germany, as there I'm using native IPv6. 1279200569 M * Pazzo (it would probably also cause issues once you try to contact other v6-hosts in the same subnet NOT sitting on the same host as long as you didn't add a global IPv6 address with a correct mask for the same subnet) 1279200613 M * Pazzo I did not (yet) play with link-local addresses from inside a vServer - routing table looks pretty "interesting" :) Did anyone already do so? 1279200692 M * Pazzo (example: ping6 to a link-local address requires you to specify the correct interface. I'm pretty curious whether this will work from inside a linux-vserver) 1279200759 M * Pazzo c) interface names in ipv6 routing table -> it shows interfaces named if4 and if10 - is this somehow related to how isolation is done by linux-vserver? 1279200799 M * daniel_hozac yes 1279200823 M * daniel_hozac interfaces which aren't visible in the guest will be shown as ifX, where X is the interface id. 1279200913 M * Pazzo daniel_hozac: thanks! Didn't even realize that this was done this way also for IPv4 :) 1279200967 M * Pazzo Bertl: you wrote about a "potentially incomplete ipv6 isolation" in one of your IPv6-related mails on the vserver-ML... incomplete in what sense? 1279201010 M * Pazzo danielhozac: btw, do you have an explaination for (b)? 1279201088 M * daniel_hozac i would have to look at the code. 1279201114 M * Pazzo Did you try out the IPv6 support? 1279201170 M * daniel_hozac i've been using IPv6 since we added it, yes. 1279201182 M * daniel_hozac even before then. 1279201188 M * Pazzo And does it respect your masks in your setup? 1279201199 M * Pazzo Or does it also "opt" for /128? 1279201225 M * daniel_hozac i haven't given any guest an entire network. 1279201251 M * Pazzo "given"? 1279201278 M * daniel_hozac my guests only have one IP each 1279201279 M * Pazzo I don't "give" an IPv4 vserver a /24 if I configure 192.0.2.17/24, do I? 1279201284 M * daniel_hozac no. 1279201309 M * daniel_hozac so your issue is that the IP assignment ends up being /128? 1279201318 M * Pazzo yes 1279201342 M * daniel_hozac and prefix is set to 64? 1279201396 M * Pazzo "prefix" is 2001:db8:some:thing, "mask" is 64 1279201415 M * Pazzo interfaces/1/mask -> 64 1279201421 M * daniel_hozac that's wrong. 1279201424 M * daniel_hozac mask is a netmask. 1279201442 M * daniel_hozac which would be ffff:ffff:ffff:ffff::, if IPv6 was supported there. 1279201452 M * Pazzo ?? 1279201463 M * daniel_hozac you should set interfaces/1/prefix to 64. 1279201467 M * Pazzo ffff:ffff:ffff:ffff:: = /64, isn't it? 1279201473 M * daniel_hozac yes. 1279201480 M * daniel_hozac that is what i said. 1279201491 M * Pazzo prefix is something like 2001:db8:: in IPv6 speech 1279201498 M * daniel_hozac yes, i know. 1279201512 A * Pazzo is confused 1279201534 M * Pazzo So why should I set interfaces/1/prefix to 64?? 1279201547 M * daniel_hozac because that's how bits designate the prefix. 1279201584 M * Pazzo http://linux-vserver.org/IPv6 -> "There is no need to add the prefix into /etc/vservers//interfaces/X/prefix. " 1279201614 M * daniel_hozac well, whoever wrote that obviously didn't know :-) 1279201646 M * Pazzo Please forgive me if I'm insisting, I still don't get it :p 1279201665 M * Pazzo So: what has 64 have to do with the prefix?? 1279201669 M * Bertl you simply can't use 'mask' 1279201674 M * Bertl use prefix instead 1279201677 M * Pazzo A prefix is a prefix, a mask is a mask 1279201680 M * daniel_hozac since, you know, it's a mask. 1279201680 M * Pazzo Hi Bertl!! 1279201683 M * daniel_hozac it's not a mask. 1279201688 M * daniel_hozac /64 is not a mask. 1279201703 M * daniel_hozac netmasks are in IP-format. 1279201715 M * daniel_hozac /64 is not an IP. 1279201723 M * Pazzo 64 is a bitmask, ffff:: is a netmask, both are masks 1279201727 M * daniel_hozac no 1279201730 M * daniel_hozac 64 is not a bitmask 1279201738 M * Pazzo ? 1279201743 M * daniel_hozac i mean, it can be, but it would not be useful to you. 1279201747 J * manana ~mayday090@84.17.25.149 1279201749 M * Pazzo Why not? 1279201764 M * Pazzo A prefix and a bitmask toghether define your IPv6 subnet... 1279201767 M * daniel_hozac because that would be equivalent to ::20 1279201776 M * Pazzo ...like a network address and a bitmask did in IPv4 1279201783 M * daniel_hozac it's not a bitmask. 1279201801 M * daniel_hozac just read the great flower page. 1279201831 M * Bertl Pazzo: regardless of the nomenclature you are using, the 'mask' file has to have an entry like a.b.c.d 1279201832 M * daniel_hozac prefix is the number of bits in the prefix. 1279201838 M * daniel_hozac i.e. a /64. 1279201860 M * Bertl Pazzo: and ipv6 addresses do not use that file, they use the 'prefix' file instead 1279201863 A * Pazzo is even more confused 1279201871 M * Pazzo Bertl: prefix file makes sens 1279201873 M * Pazzo (e) 1279201887 M * daniel_hozac look 1279201890 M * Pazzo But: a prefix is something like "2001::db8::/64" 1279201890 M * daniel_hozac this all predates IPv6 1279201928 M * daniel_hozac the /xx is a CIDR-prefix. 1279201929 M * daniel_hozac thus 1279201933 M * Bertl in the Linux-VServer case, the prefix is 'just' the 64 part 1279201938 M * daniel_hozac /xx goes in the prefix file. 1279201974 M * Pazzo Let's forget about the naming, I resign ;-) So I have to use the prefix file, I'm fine with that 1279201986 M * Pazzo Does it have to contain /64 or just "64"? 1279202002 M * daniel_hozac 64 1279202031 M * Pazzo daniel_hozac: works perfectly, (b) is resolved :) 1279202038 M * Pazzo Thank you for your patience! 1279202062 M * Bertl np, regarding the incomplete part, bonbon is the ipv6 'expert' here 1279202066 M * Pazzo (someone should add a note to http://linux-vserver.org/IPv6 ) 1279202078 M * Bertl please do so :) 1279202113 M * Pazzo May I remove "There is no need to add the prefix into /etc/vservers//interfaces/X/prefix." 1279202123 M * daniel_hozac yes 1279202126 M * daniel_hozac since that's just wrong. 1279202443 M * Pazzo ok. Should I add a note related to "prefix"? It works exactly as it did in IPv4 (for linux-vserver), but as you noted the word "prefix" may confuse people once it goes to IPv6 ;-) 1279202496 M * daniel_hozac if you want 1279202601 M * Pazzo Just to deeply understand it (and to not write any rubbish): util-vserver allows you to use either "mask" (255.255.255.0 or ffff:ffff:ffff:ffff::) or "prefix" (/24 or /64), correct? 1279202706 M * Pazzo And: is there a way to get a ::1/128 on your lo? (127.0.0.1/8 is already there as of nflags (HIDE_LBACK, LBACK_REMAP) 1279203227 M * daniel_hozac no 1279203231 M * daniel_hozac there is no way to use mask. 1279203249 M * daniel_hozac and there is no loopback isolation for IPv6. 1279203709 Q * manana Remote host closed the connection 1279204210 J * manana ~mayday090@84.17.25.149 1279204812 Q * manana Ping timeout: 480 seconds 1279204845 J * manana ~mayday090@84.17.25.149 1279205241 M * Pazzo daniel_hozac: could you check whether what I have written to /IPv6 is correct? (been unsure whether to write SHOULD or MUST :p) 1279205502 Q * manana Read error: Connection reset by peer 1279205535 J * manana ~mayday090@84.17.25.149 1279205564 J * dowdle ~dowdle@scott.coe.montana.edu 1279208073 J * imcsk8 ~ichavero@evdomip-67-179.iusacell.net 1279209286 J * dna ~dna@76-198-103-86.dynamic.dsl.tng.de 1279210336 J * bonbons ~bonbons@2001:960:7ab:0:2c0:9fff:fe2d:39d 1279211873 J * thierryp ~thierry@zankai.inria.fr 1279212836 Q * petzsch Read error: Connection reset by peer 1279213550 J * petzsch ~markus@dslb-092-078-157-225.pools.arcor-ip.net 1279213716 J * harobed ~harobed@arl57-1-82-231-110-14.fbx.proxad.net 1279214067 Q * imcsk8 Quit: This computer has gone to sleep 1279214627 Q * dna Quit: Verlassend 1279214765 Q * Pazzo Quit: Bye! 1279215388 Q * ntrs_ Ping timeout: 480 seconds 1279216289 Q * gnuk Ping timeout: 480 seconds 1279216774 Q * thierryp Ping timeout: 480 seconds 1279217700 Q * derjohn_foo Ping timeout: 480 seconds 1279219457 Q * fLoo Remote host closed the connection 1279219570 Q * petzsch Read error: Connection reset by peer 1279219909 Q * fback Quit: przeprowadzka | relocating 1279221149 J * thierryp ~thierry@home.parmentelat.net 1279222478 J * skoolkid ~skoolkid@63.81.0.20 1279223153 M * skoolkid anyone are able to run ssh on the guest? 1279223180 M * daniel_hozac yes 1279223184 M * daniel_hozac you just have to read the FAQ first. 1279223211 Q * Radiance Ping timeout: 480 seconds 1279223645 M * ard bah 1279223658 A * ard hates when somebody does not configure serial console properly :-( 1279223716 M * ard daniel_hozac : got one kernel issue, but I guess that should be fixed with: http://patchwork.ozlabs.org/patch/53098/ 1279223731 J * imcsk8 ~ichavero@evdomip-67-179.iusacell.net 1279223735 Q * thierryp Remote host closed the connection 1279223749 M * ard but for now my system is in a reboot mode so I can't test if that's the ultimate fix. 1279223764 M * ard (hangs in reboot, and serial console was not configured etc...) 1279223781 M * ard ow wait... ipmi power reset 1279224276 M * ard So, forget that fix... 1279224284 M * ard I now even get a: 1279224286 M * ard Jul 15 22:01:57 imperium kernel: sysfs: cannot create duplicate filename '/devices/virtual/net/lo' 1279224310 M * ard before that it was just: 1279224311 M * ard Jul 15 20:00:20 imperium kernel: sysfs: cannot create duplicate filename '/devices/virtual/net/vlan2941/statistics' 1279224432 J * dejan ~dejan@nimitz.rosehosting.com 1279224589 J * ntrs ~ntrs@77.28.167.44 1279225036 M * dejan Hi, I was using Linux 2.6.22.19 with Linux-VServer 2.2.0.7. and I set processes limits with '/etc/vservers//rlimits/nproc' . Now I'm using Linux 2.6.33.5 with Linux-VServer 2.3.0.36.30.4 . I do not know how to set processes limits. Please advise. 1279225147 M * daniel_hozac same file. 1279225265 M * dejan Thanks. 1279225311 J * derjohn_foo ~aj@d065052.adsl.hansenet.de 1279225449 Q * dejan 1279225559 Q * micah Read error: Operation timed out 1279225682 Q * dannf Ping timeout: 480 seconds 1279225732 J * dejan ~dejan@nimitz.rosehosting.com 1279225748 Q * dejan 1279226595 J * ktwilight_ ~keliew@91.176.187.5 1279226933 Q * bonbons Quit: Leaving 1279227308 J * petzsch ~markus@dslb-092-078-157-225.pools.arcor-ip.net 1279228613 Q * petzsch Quit: Leaving. 1279229020 Q * ntrs Ping timeout: 480 seconds 1279230687 J * micah ~micah@micah.riseup.net 1279230886 J * dannf ~dannf@utter.lackof.org 1279231186 Q * harobed Ping timeout: 480 seconds 1279232424 M * daniel_hozac ard: did the change to util-vserver work out? 1279232815 P * skoolkid thanks for the help daniel_hozac 1279232960 J * thierryp ~thierry@home.parmentelat.net 1279232994 Q * thierryp Remote host closed the connection 1279233167 M * ard daniel_hozac : with some minor tweaks yes... 1279233175 M * ard Now for some kernelbug hunting 1279233207 M * ard the debian/ packaging is finished except for the vprocunhide, which I think should be run from /etc/init.d/util-vserver 1279233314 M * daniel_hozac oh? 1279233318 M * daniel_hozac do you have a tar? 1279233337 M * daniel_hozac vprocunhide is separate by design. 1279233346 M * daniel_hozac its config may change and you may want to rerun it. 1279233360 M * daniel_hozac rerunning the util-vserver initscript on a live system would stop all your guests. 1279233382 M * daniel_hozac i have started packaging it as well, i noticed that my Debian boxes are seriously out of date. 1279233673 M * ard file upload of firefox is amazingly buggy 1279233683 M * ard when my ~ is too big, it refuses to select the right file 1279233708 M * daniel_hozac scp ftw :-) 1279233761 M * ard https://projects.kwaak.net/twiki/pub/Misc/VserverMisc/util-vserver-debian.tar.gz 1279233766 M * ard or minus the s 1279233783 M * ard 2.3 K :-) 1279233793 M * ard it's based on micah's work 1279233800 M * ard removed al kinds of patching 1279233825 M * ard so in a sense it's almost the original install.sh script :-) 1279233839 M * daniel_hozac does it have the separate packages? 1279233845 M * ard no :-( 1279233859 M * ard it's one util-vserver 1279233890 M * ard you want it like the .spec file? 1279233902 M * daniel_hozac yes, that is what i'm working on. 1279234009 M * ard hmmm lib-util-vserver-dev :-) 1279234046 M * ard It's possible, but it takes some more work... 1279234050 M * daniel_hozac well, mine is calling it libvserver0-dev 1279234085 M * ard Hmmm, that's sounds better 1279234265 A * ard can't find libvserver0-dev in the spec file 1279234273 M * ard or do you mean the debian packages? 1279234286 M * daniel_hozac the debian package. 1279234290 M * ard ok :-) 1279234325 M * ard I can look at your debian tree, or better yet, I let a colleage do that (an official debian maintainer) 1279234345 A * ard is going to get some food 1279234395 M * daniel_hozac i will have to get back to that later. 1279235121 Q * dowdle Remote host closed the connection 1279236274 Q * imcsk8 Quit: This computer has gone to sleep