[00:00] I'm doing a small text which I'll place on my webserver where I describe the thing shortely. when I have time I'll get into more detail. [00:01] the vproc stuff? [00:01] or what? [00:02] Nick change: Doener_ -> Doener [00:02] yep [00:02] add it to the wiki :) [00:02] exactly what I wanted to suggest ... [00:02] right! thanks. I think I saw it while searching for that problem. [00:02] oh, sorry... [00:03] http://www.linux-vserver.org/index.php?page=Proc-Security [00:03] netrose (~netrose@FL3-24.217.241.239.charter-stl.com) left irc: Read error: Connection reset by peer [00:10] ExpiryJames (~james@h24-71-63-164.ok.shawcable.net) joined #vserver. [00:11] hi James! [00:11] Bertl: Hi. [00:12] Got a minute for a bit of off-topic? :) [00:13] sure ... [00:14] I'm compiling a cloop module (compressed loop from knoppix) against 2.6 kernel and get a strange error while loading it: -1 Unknown symbol in module [00:15] that isn't that strange, is it? [00:15] Even more, on a neighbour's box it states something like -1 Invalid module format. [00:16] Actually that's not clear to me. [00:16] this means too old modutils ... [00:16] Hm. [00:16] But other modules load quite fine. [00:16] do you have magic sysreq enabled? [00:16] is there a /proc/sysrq-trigger ? [00:17] Yes, it's there. [00:17] okay do echo 9 > /proc/sysrq-trigger [00:17] Done. [00:17] the try to load the module, and have a look at the kernel log (dmesg) [00:18] cloop: Unknown symbol ntohl [00:18] well, there you go [00:18] ntohl probably a missing/bad include ... [00:18] Hm. Cool. :) [00:27] Bertl: Look, it is defined like that: [00:27] #define ntohl(x) ___ntohl(x) [00:27] hmm, yes, and? [00:27] Inside a ifdef/endif block [00:28] This turns to that I need to recompile kernel with those defines.. [00:28] unlikely ... [00:29] Why? [00:29] try to find a .c source file, which uses ntohl() [00:29] look what headers are included, and include them ... [00:30] gotta boot linux! cul8er [00:30] cu [00:31] fleshcrawler (~fleshcraw@port-212-202-204-54.reverse.qdsl-home.de) left irc: [00:32] Hm. You were right. :) [00:33] hmm, correct again! common guys, group hug! [00:34] TheSeer (~theseer@border.office.salesemotion.net) joined #vserver. [00:44] hi Seer! [00:49] hey all [00:50] Is there a nice vserver way to force a shutdown of a vserver [00:51] which vserver release? [00:51] right now I have a vserver that hangs shutting down eth0 [00:51] latest [00:51] kernel 1.26 [00:52] Nick change: cgone -> cdub [00:52] okay, do you want to know how to kill/remove it or do you want to know how to fix the shutdown issue? [00:52] lets try for fixing the problem// [00:53] I started up the vserver with IP that wer assigned to the root server [00:53] okay, which distro do you use inside the vserver? [00:53] RH7.X [00:53] root =RH9.0 [00:53] Pompom... [00:53] *hum* [00:54] I removed and downed the ip's in the root server [00:54] ExpiryJames: okay, check the runlevel scripts in /vservers//etc/rc3.d/ [00:54] *wave* [00:54] hi click! [00:55] heya [00:55] :)) [00:56] so /vservers/xxxx.sh ? [00:56] no, the runlevel scripts inside the vserver installation [00:56] k, [00:56] miss understood [00:57] k, in that directory [00:57] you probably have some start/stop scripts there, which try to init/shutdown the interface ... [00:58] S10network -> ../init.d/network [00:58] and a K*network probably too ... [00:58] change that to K10network -> ../init.d/network [00:59] okay, you can safely remove those links ... [00:59] same goes for other services which won't work inside a vserver, like hardware detection or actual reboot ... [00:59] this can also be done via chkconfig in RH IIRC [01:02] k [01:04] still sticking on network [01:05] hmm, could you provide a log? [01:05] from within the vserver? [01:06] no, from the failing/hanging shutdown ... [01:07] I am ctrl+c ing it from the root server [01:08] it went down now [01:08] yeah, okay, just show me the lines ... [01:09] .com /vservers/stepsites/etc/rc3.d]# vserver stepsites stop [01:09] Stopping the virtual server stepsites [01:09] Server stepsites is running [01:09] ipv4root is now 69.59.181.213 69.59.157.213 69.59.181.214 [01:09] New security context is 213 [01:09] Shutting down smartd: [FAILED] [01:09] Shutting down interface eth0: [01:09] ctrl+c [01:09] this means that the K* script is still active ... [01:10] trying to shut down the network interface from inside the vserver, which isn't possible ... [01:10] that was before... it now is down,, now longer is listed in vserver stat [01:10] which is where it was hanging.. [01:10] ic [01:26] iprone (~bob_uriah@216.89.233.129) joined #vserver. [01:26] hi iprone! [01:27] hi...just checking out the list, been using vserver for quite a while [01:27] yeah, and do you like it? [01:28] definitely, some of the guys here are trying to figure out how you have so much time/patience, but we definitely appreciate the work you guys have do some far [01:28] im actually from eapps, we are setting up an old alpha so that you can do some testing on [01:29] sounds good ... [01:30] i think youve been talking with Rick Lingsch about it, havent been in the loop up until now actually [01:31] yeah, was exchainging emails with him ... [01:31] and with Kyle ... [01:32] so what do you want to know about me or vserver? [01:33] oh...nothing specific, i was looking in, thinking maybe i could jump in if someone had some questions or needed some help [01:33] just trying to help out as much as possible since ive been leeching from the project for so long [01:33] hey I like that ... [01:34] ensc (~ircensc@ultra.csn.tu-chemnitz.de) left irc: Ping timeout: 480 seconds [01:35] so my last question to Kyle was, if there would be anyone at your site who could keep an eye on the alpha, maybe even doing some vserver alpha testing? [01:35] yeah definitely [01:35] because that is actually something I started worrying about ... [01:35] ok [01:37] btw, did you enter your company to the wiki yet? [01:37] no not yet [01:37] (well, it's up to you, so only do it if you feel like ;) [01:38] yeah, we were talking about doing it, we will definitely [01:40] btw, we managed to get the 0.09 (2.6.3.) working on x86_64 today ... [01:44] im interested to see how that is going to work, we just started setting up a 2.6.3-0.09 test server internally, to check the new functionality [01:45] based on ia32 or ia32e? [01:45] or amd64? [01:45] on ia-32...but when doing testing, do you have a specific sequence of things that are required to be done [01:45] to have a standard reporting back to the group? [01:46] nope, not yet, but if you think that would be useful, go ahead create a standard ... [01:46] basically ml is a good choice atm [01:47] ok [01:47] we are thinking about regression tests ... but besides my simple script (testme.sh) nothing has been done yet [01:48] ensc (~ircensc@ultra.csn.tu-chemnitz.de) joined #vserver. [01:49] ah, enrico, good morning! [01:49] just the same, old story named "BNC network" [01:50] so I read you finished the vcontext? [01:50] yep, is in CVS [01:51] btw, is the vkill now in util-vserver a/b/c? [01:51] should be in stable too [01:51] ah okay ... so I can forget about the vkill tar gz, right? [01:51] yep [01:51] good, thanks! [01:59] hmm, just finished adapting the TB scheduler to 2.6.3/vs0.09++, anybody interested in testing? [02:08] hi [02:08] cannot run 2.6 as of now... [02:08] :( [02:08] still no luck with vmware? [02:08] nope [02:09] why not use QEMU? [02:09] i could try it [02:09] never used it [02:09] well, it works with 2.6.3 ;) [02:10] is the integrated tb scheduler patch vs0.09.3? [02:10] nope ... didn't release it yet ... [02:10] you want to try it? [02:11] netrose (~netrose@FL3-24.217.241.239.charter-stl.com) joined #vserver. [02:11] hi bobi! [02:11] definitely [02:11] okay, just a second ... [02:13] JonB (~NoSuchUse@129.142.112.33.ip.tele2adsl.dk) left irc: Quit: Leaving [02:13] http://vserver.13thfloor.at/Experimental/delta-2.6.3-vs0.09-vs0.09.4.diff [02:13] netrose (~netrose@FL3-24.217.241.239.charter-stl.com) left irc: Remote host closed the connection [02:14] netrose (john877@FL3-24.217.241.239.charter-stl.com) joined #vserver. [02:14] but the controlling syscall and the /proc interface isn't available yet ... [02:14] god damn i wish hpt would write a driver for this card [02:15] matta, IIRC, andre hedrick was working on a SATA driver for 2.6 ... [02:16] for this card? [02:16] don't know ... [02:17] tell you the truth, i don't even want this card [02:17] look at this [02:17] http://www.amdzone.com/articleview.cfm?articleid=1362&page=4 [02:18] yeah and, this is it? [02:18] RocketRAID 1640 ? [02:19] yeah [02:19] i'm not finding anyone who likes this card, i'd rather pay a few extra bucks a month and have this place use 3ware cards [02:19] that would be a good choice anyway ... [02:22] I was able to get 32 ip addresses using the util-vserver tools. Below is a diff of the util-vserver-0.29 tools for reference. I used 64 instead of 32 because chbind would error out when trying to bind 32 ip addresses with the lower setting. I am assuming that the tools possibly added the loopback address as one of the 32 allowed. If anyone can shed some light on that, would be appreciated. Otherwise, the diff below should work. [02:23] Also, to be consistent, we changed NB_IPV4ROOT to 64 in the kernel. Although we havent tested this, its possible the tools can now support up to 63 ips. [02:23] is there really demand for that? and if wouldn't address ranges be a better choice? [02:23] (just collecting opinions ;) [02:24] i actually wrote that [02:24] such fixed array-sizes are broken by design.... [02:25] our subnets are in disarray from being used for so long, that ranges wouldnt work very well [02:25] agreed, but a generic solution isn't that easy ... [02:25] well, actually it's easy, but not performant ... [02:26] netrose (john877@FL3-24.217.241.239.charter-stl.com) left irc: Ping timeout: 480 seconds [02:26] so what is the rationale behind using so many IPs on one vserver? [02:27] (where so many isn't 16 actually ;) [02:29] morning [02:29] yeah, well in our setup, we have 32 customers in each vserver, and we allow an ip address for each customer [02:29] morning kestrelw! [02:29] heh [02:30] 32 per vserver, keen [02:30] kevin?: so one ip for SSL? [02:30] yes [02:31] okay, that makes sense ... [02:31] hey, i have a question...if you have two vservers with exactly the same shared libraries (libc for example), but the volumes are not unified...are the shared memory libraries still shared across vservers or is each library loaded once for each vserver? [02:31] Bertl: if there was a virtualized network stack this wouldn't bw an issue :) [02:31] iprone: but why, for example not allocating an entire subnet to a vserver? [02:31] matta: it would be ... [02:32] how would it? [02:32] you're saying the range? [02:32] we could do that, but we have various hosting plans, not just in vserver, we would be tying up resources that could be used elsewhere [02:32] you could use routing to simplify that ... [02:33] kestrelw: it's shared if it's the same inode [02:33] for example assigning the entire subnet to a vserver, and just routing the used ips [02:34] kestrelw: the inode cache is shared, the vm copy RO can be shared ... [02:34] we could do that, but but we want to make our plans as mobile as possible, for example, lets say we move one of the 32 customers from inside that vserver to their own, we would then need to give them a new ip address [02:34] jes (~jes@cpc1-leed5-3-0-cust196.ldst.cable.ntl.com) joined #vserver. [02:34] evening all [02:34] evening jes! [02:35] heya Bertl...and how are you? [02:35] okay, thanks [02:35] still fine ;) [02:35] I just added TB-sched support to 2.6.3-vs0.09.4 [02:36] ahhh excellent [02:36] hiya kestrelw [02:36] well I'm still puzzling over my tools no longer rebuilding [02:36] done a completely fresh kernel recompile [02:36] still no dice [02:36] ;) [02:36] ensc is here ... [02:36] *hint* [02:37] *poke* ensc [02:37] lol [02:37] jes: which error? [02:37] to cut a long story short ensc, all was working fine earlier [02:37] then I decided to recompile the tools so I could change the rootdir [02:37] now it no longer recompiles...errors such as - [02:37] try 'rm -f config.cache' [02:38] or compile in new source tree [02:38] In file included from /lib/modules/2.6.3/build/include/linux/thread_info.h:20, [02:38] from /lib/modules/2.6.3/build/include/linux/spinlock.h:12, [02:38] from /lib/modules/2.6.3/build/include/linux/blockgroup_lock.h:8, [02:38] from /lib/modules/2.6.3/build/include/linux/ext2_fs_sb.h:19, [02:38] from /lib/modules/2.6.3/build/include/linux/ext2_fs.h:20, [02:38] from lib/ext2fs.h:25, [02:38] from lib/ioctl-getext2flags.hc:24, [02:38] from lib/syscall_getiattr-fscompat.hc:23, [02:38] from lib/syscall_getiattr.c:36: [02:38] /lib/modules/2.6.3/build/include/linux/bitops.h: In function `get_bitmask_order': [02:38] /lib/modules/2.6.3/build/include/linux/bitops.h:78: warning: implicit declaration of function `fls' [02:38] ending with a - [02:39] /lib/modules/2.6.3/build/include/asm/io_apic.h:66: warning: no semicolon at end of struct or union [02:39] /lib/modules/2.6.3/build/include/asm/io_apic.h:68: error: parse error before "u32" [02:39] /lib/modules/2.6.3/build/include/asm/io_apic.h:68: warning: no semicolon at end of struct or union [02:39] /lib/modules/2.6.3/build/include/asm/io_apic.h:73: warning: type defaults to `int' in declaration of `bits' [02:39] /lib/modules/2.6.3/build/include/asm/io_apic.h:73: warning: `packed' attribute ignored [02:39] /lib/modules/2.6.3/build/include/asm/io_apic.h:73: confused by earlier errors, bailing out [02:39] make[2]: *** [lib/lib_libvserver_la-syscall_getiattr.lo] Error 1 [02:39] make[2]: Leaving directory `/home/jes/tools/vserver/util-vserver-0.29.196' [02:39] make[1]: *** [all-recursive] Error 1 [02:39] make[1]: Leaving directory `/home/jes/tools/vserver/util-vserver-0.29.196' [02:39] make: *** [all] Error 2 [02:39] there's no config.cache file [02:39] BTW there's lots more warnings/errors between those two pastes [02:39] try to install e2fsprogs-devel [02:40] libe2fs-devel [02:40] ok one sec [02:41] I do not know why it worked before; ./configure should detect broken kernel headers, but probably there are some previous #include's in the source which are making it fail there but succeed in ./configure [02:41] libext2fs2-devel [02:41] trying to find the .tgz [02:41] e2fslibs-dev on debian ;) [02:41] jes, you are using mandrake, just do [02:41] urpmi libext2fs2-devel [02:42] ty Bertl [02:44] wow...that did it [02:44] ;) [02:44] ty ensc, ty Bertl [02:45] ensc: hmm, so util-vserver looks first in the glibc, then in libext3, then in the kernel tree, right? [02:45] BTW I'm guessing I should be compiling with dietlibc..considering the warning ./configure gives me [02:47] Bertl: no, and then [02:47] there should be no -I [02:48] iprone: could you initiate an opinion poll, how many IPs (as an average) would be sufficient for one vserver, and whether ranges would be a better choice than separate IPs? [02:49] sure can [02:49] ah, I know. While compiling the library, I use '-I' which includes other headers than these being used by ./configure [02:51] hmmmm [02:51] I just installed dietlibc-devel [02:51] try the compile now and it gets 99.9999% through [02:51] but throws up - [02:51] diet gcc -g -O2 -std=c99 -Wall -pedantic -W -o src/reducecap src/reducecap.o lib/.libs/libvserver.a [02:51] src/reducecap.o(.text+0x255): In function `show': [02:51] src/reducecap.c:202: undefined reference to `capget' [02:51] collect2: ld returned 1 exit status [02:51] make[2]: *** [src/reducecap] Error 1 [02:52] jes: which dietlibc version? [02:52] one sec [02:52] 0.22-7 [02:53] [root@cerberus util-vserver-0.29.196]# rpm -qa | grep diet [02:53] dietlibc-devel-0.22-7mdk [02:53] a little bit old... any output at 'grep -r capget /usr/lib/dietlibc/' [02:53] (or /usr/lib64) [02:53] [root@cerberus util-vserver-0.29.196]# grep -r capget /usr/lib/dietlibc/ [02:53] grep: /usr/lib/dietlibc/: No such file or directory [02:54] 0.24: [02:54] Olaf: added missing syscalls (cap{g,s}et) and addmntent [02:54] still grepping the lib64 [02:55] [root@cerberus util-vserver-0.29.196]# grep -r capget /usr/lib64 [02:55] /usr/lib64/32/perl5/5.8.1/x86_64-linux-thread-multi/asm-i386/unistd.ph: eval 'sub __NR_capget () {184;}' unless defined(&__NR_capget); [02:55] /usr/lib64/32/perl5/5.8.1/x86_64-linux-thread-multi/asm-x86_64/unistd.ph: eval 'sub __NR_capget () {125;}' unless defined(&__NR_capget); [02:55] /usr/lib64/32/perl5/5.8.1/x86_64-linux-thread-multi/bits/syscall.ph:unless(defined(&SYS_capget)) { [02:55] /usr/lib64/32/perl5/5.8.1/x86_64-linux-thread-multi/bits/syscall.ph: sub SYS_capget () { &__NR_capget;} [02:55] ... only /usr/lib64/dietlibc ;) [02:55] lol thats all it found [02:55] it is expected that it fails for dietlibc <0.24 [02:56] ahhh ok [02:56] upgrade time ;) [02:56] where can I get a recen dietlibc? [02:56] http://www.fefe.de/dietlibc/ [02:57] yuk...no ./configure script with it ;) [02:58] is mandrake patching it too heavy that it would be too difficultly to adapt the old .spec file? [02:59] I got the .tgz [03:02] ick, this diet install it crap...it puts stuff in very strange places [03:03] http://vserver.13thfloor.at/Stuff/dietlibc-0.24-1mdk.src.rpm [03:04] ty Bertl [03:04] x86_64/sigaction.c:32: warning: redundant redeclaration of `__rt_sigaction' in same scope [03:04] x86_64/sigaction.c:6: warning: previous declaration of `__rt_sigaction' [03:04] x86_64/sigaction.c:34: error: redefinition of `restore_rt' [03:04] x86_64/sigaction.c:8: error: `restore_rt' previously defined here [03:04] x86_64/sigaction.c:38: warning: redundant redeclaration of `__libc_sigaction' in same scope [03:04] x86_64/sigaction.c:13: warning: previous declaration of `__libc_sigaction' [03:04] x86_64/sigaction.c:39: error: redefinition of `__libc_sigaction' [03:04] x86_64/sigaction.c:13: error: `__libc_sigaction' previously defined here [03:05] x86_64/sigaction.c:52: error: redefinition of `sigaction' [03:05] x86_64/sigaction.c:26: error: `sigaction' previously defined here [03:05] make: *** [bin-x86_64/sigaction.o] Error 1 [03:05] error: Bad exit status from /var/tmp/rpm-tmp.77202 (%build) [03:05] RPM build errors: [03:05] Bad exit status from /var/tmp/rpm-tmp.77202 (%build) [03:05] lol not my night tonight [03:05] hmm, okay do an rpm -i [03:05] of the src rpm [03:05] and comment out the lines stating with %patch [03:06] k [03:06] in the spec file [03:06] whats rpm -i do? [03:06] it didn't seem to return anything [03:07] it installs the sources ... [03:07] okay follow me: [03:07] rpm -i dietlibc-0.24-1mdk.src.rpm [03:07] [root@cerberus tmp]# rpm -i dietlibc-0.24-1mdk.src.rpm [03:07] [root@cerberus tmp]# [03:07] thats what I got [03:07] vi /usr/src/RPM/SPECS/dietlibc.spec [03:07] no .spec file [03:08] vi /usr/src/RPM/SPECS/dietlibc.spec [03:08] comment out the following lines ... [03:08] %patch0 -p1 -b .features [03:08] %patch1 -p1 -b .config [03:08] %patch2 -p1 -b .ppc [03:08] %patch3 -p1 -b .x86_64-sigaction [03:08] ahhh ok [03:08] %patch5 -p1 -b .tests [03:08] %patch7 -p1 -b .lib64 [03:08] %patch8 -p1 -b .net-ethernet [03:09] probably only %patch3 is already included ... [03:09] % is comment in spec files? [03:09] or # ? [03:09] # [03:09] ok, nothing is commented out [03:10] ok done [03:10] rpm -ba /usr/src/RPM/SPECS/dietlibc.spec [03:11] RPM build errors: [03:11] File not found: /var/tmp/dietlibc-0.24-buildroot/usr/bin/diet [03:11] File not found: /var/tmp/dietlibc-0.24-buildroot/usr/lib64/dietlibc [03:11] File not found by glob: /var/tmp/dietlibc-0.24-buildroot/usr/share/man/man*/* [03:11] it looks ok up to there [03:11] okay, for a test, leave %patch3 commented out, and uncomment the others ... [03:13] hmmmm [03:13] bit better [03:13] but get a few lines like - [03:13] shell-init: could not get current directory: getcwd: cannot access parent directories: No such file or directory [03:13] + umask 022 [03:13] + cd /usr/src/RPM/BUILD [03:13] chdir: could not get current directory: getcwd: cannot access parent directories: No such file or directory [03:13] that is because you entered the build dir ;) [03:13] gah! [03:14] ok MUCh better [03:14] ;) [03:14] done [03:16] now install the rpm? [03:16] yup! [03:17] building tools now.... [03:17] In file included from /usr/lib64/dietlibc/include/asm/sigcontext.h:13, [03:17] from /usr/lib64/dietlibc/include/sys/ucontext.h:4, [03:17] from /usr/lib64/dietlibc/include/signal.h:512, [03:17] from src/pipe-sync.c:26: [03:17] /usr/lib64/dietlibc/include/asm/x86_64-sigcontext.h:11: error: parse error before "__u64" [03:17] /usr/lib64/dietlibc/include/asm/x86_64-sigcontext.h:11: warning: no semicolon at end of struct or union [03:17] /usr/lib64/dietlibc/include/asm/x86_64-sigcontext.h:12: warning: type defaults to `int' in declaration of `rdp' [03:17] /usr/lib64/dietlibc/include/asm/x86_64-sigcontext.h:12: error: ISO C forbids data definition with no type or storage class [03:17] /usr/lib64/dietlibc/include/asm/x86_64-sigcontext.h:18: error: parse error before '}' token [03:17] /usr/lib64/dietlibc/include/asm/x86_64-sigcontext.h:18: warning: ISO C does not allow extra `;' outside of a function [03:17] make[2]: *** [src/pipe-sync.o] Error 1 [03:17] heh ;) [03:17] one step forward 0.75 steps back [03:18] maybe I shouldn't go the diet route just yet? [03:18] probably ... [03:19] or ask on the maillist, or look into CVS if it solves your problem [03:19] (dietlibc maillist) [03:20] *nods* [03:20] ty anyway [03:28] wow...snowing heavy here now [03:28] iprone (~bob_uriah@216.89.233.129) left irc: Quit: [03:31] paul (~irssi@195.202.59.150) left irc: Quit: leaving [03:45] I'm cracking up again... [03:46] [root@cerberus /]# vserver skeleton build [03:46] No build-method specified [03:46] I'm sure that worked before? [03:46] yeah, but for alpha tools you need some additional args ... [03:47] ahhhh....for example? [03:50] What's an example build command with alpha tools? [03:50] have a look at the alpha utils page .. [03:50] or let enrico explain it to you ... [03:50] thanks Bert for the help today, [03:50] np [03:51] good night [03:51] ExpiryJames (~james@h24-71-63-164.ok.shawcable.net) left irc: Quit: Leaving [03:51] jes: syntax is 'vserver build -m skeleton *' [03:52] ahhh ty ensc [03:52] whats the link for the alpha utils page Bertl? [03:53] http://www.linux-vserver.org/index.php?page=alpha+util-vserver [03:53] ty [03:57] ok...thats the last question I'll be asking for a while ;) [03:57] I'm sure you've had more than enough from me ;) [03:57] well I'm okay with your questions ... [03:58] ty Bertl [04:03] hmm, did I miss the question? [04:04] lol no! [04:04] I was saying I wouldn't be asking any more, since I must have asked a million in the last couple of days ;) [04:04] okay so that's is actually that was, right? [04:06] lol yes [04:06] sorry...my bad grammar [04:07] so what about another question for free? [04:07] lol [04:07] ok... [04:07] Action: Bertl already prepared to answer another so ... [04:07] when I was tinkering earlier...I noticed that in my vserver I can see all the network connections etc being made on the host....any way to stop that? [04:08] e.g. a "netstat -tuan" on the vserver shows the ip of the host [04:08] hmm, that is probably a bug, nobody reported yet ;) [04:08] or maybe just the way I setup mine ;) [04:09] might be, but a vserver should only report the conenctions _and_ addresses it can see/bind to ... [04:10] but I'll check that in a few minutes ... might be a 2.6.x issue ... [04:11] ok ty [04:13] yeah, funny, that separation seems to be missing, it seems nobody found out yet (except for you) ... [04:13] hmmm [04:14] [root@cerberus util-vserver-0.29.196]# vserver-stat [04:14] open(): No such file or directory [04:14] hmmmm I seem to have broken something again (for a change) [04:14] ;) [04:14] yes, procfs ;) [04:15] that is the proc security again ... [04:15] ack damn! [04:15] ./vproc /proc /proc/*/ /proc/*/* ? [04:15] ./vproc -E /proc /proc/*/ /proc/*/* ? [04:15] or something like that [04:16] yeah, actually setattr --~hide /proc/* /proc/*/* /proc/*/*/* [04:17] [root@cerberus skel]# setattr --~hide /proc/* /proc/*/* /proc/*/*/* [04:17] -bash: /usr/local/sbin/setattr: Argument list too long [04:17] * now there are three bears on that island ... * [04:17] lol [04:18] setattr --~hide /proc/[a-z]* /proc/[a-z]*/* /proc/[a-z]*/*/* [04:18] ok I split it up into three different calls [04:19] netrose (john877@FL3-24.217.241.239.charter-stl.com) joined #vserver. [04:20] lol damn...just that lib64 chroot problem again now [04:20] I'll get there eventually ;) [04:22] hmmm inside the vserver now...and I get [04:22] [root@cerberus /]# w [04:22] Error: /proc must be mounted [04:22] To mount /proc at boot you need an /etc/fstab line like: [04:22] /proc /proc proc defaults [04:25] again proc security ... * four bears ... * [04:25] verify with showattr ... [04:26] ? but I just did the setattr [04:26] you should see a 'h' instead of a 'H' [04:26] showattr on where? [04:27] showattr /proc/ [04:27] (on the hsot) [04:27] s/hsot/host/ [04:27] [root@cerberus local]# showattr /proc/ [04:27] awhbi- /proc/ [04:27] Awhbi- /proc/ide [04:27] Awhbi- /proc/crypto [04:27] Awhbi- /proc/swaps [04:27] Awhbi- /proc/kallsyms [04:27] hmmm [04:28] okay, check with cat /proc/mounts ... [04:28] maybe the proc actually isn't mounted ... [04:28] [root@cerberus skel]# cat /proc/mounts [04:28] rootfs / rootfs rw 0 0 [04:28] /dev/root / ext3 rw 0 0 [04:28] none /proc proc rw 0 0 [04:28] none /sys sysfs rw 0 0 [04:28] none /dev/pts devpts rw 0 0 [04:28] /dev/hda5 /tmp ext3 rw 0 0 [04:28] /dev/hda1 /var ext3 rw 0 0 [04:28] /dev/hde1 /usr/local ext3 rw 0 0 [04:28] /dev/hdg1 /home ext3 rw 0 0 [04:28] thats on the host [04:28] hmm, okay, so the proc wasn't mounted in the vserver 8-) [04:29] ahhh...I thought the "vserver skel start" did the mount? [04:29] it should ... [04:30] try to cycle the vserver (stop/start) and if it is still missing, please ask enrico ... [04:31] hmmm yep definitely not mounting it [04:31] mounting happens in an own namespace [04:31] try 'touch /etc/vservers/.defaults/nonamespace' [04:32] hmmm I didn't have a /etc/vservers/.defaults dir [04:33] then /usr/local/etc/vservers (or wherever you installed it) [04:33] ok...in the vserver I now get - [04:34] [root@cerberus /]# mount [04:34] /dev/hdv1 on / type ext2 (defaults) [04:34] none on /proc type proc (defaults) [04:34] none on /tmp type tmpfs (size=16m,mode=1777) [04:34] none on /dev/pts type devpts (gid=5,mode=620) [04:34] [root@cerberus /]# w [04:34] Error: /proc must be mounted [04:34] To mount /proc at boot you need an /etc/fstab line like: [04:34] /proc /proc proc defaults [04:34] In the meantime, mount /proc /proc -t proc [04:35] try 'cat /proc/mounts'; this reflects the reality [04:36] [root@cerberus /]# cat /proc/mounts [04:36] cat: /proc/mounts: No such file or directory [04:36] thats in the vserver [04:36] hehe hard to do without proc ;) [04:36] lol [04:36] chicken and egg? [04:37] jes: which 'nonamespace' file do you have touched? [04:37] errr.... [04:37] I did what you told me ; [04:37] ;) [04:37] touch /etc/vservers/.defaults/nonamespace [04:38] jes: post the outputof 'vserver-info - SYSINFO | egrep 'Dire|prefix'' [04:38] ok, it works if I mount proc myself manually in the host [04:39] [root@cerberus .defaults]# vserver-info - SYSINFO | egrep 'Dire|prefix' [04:39] prefix: NONE [04:39] sysconf-Directory: ${prefix}/etc [04:39] cfg-Directory: ${prefix}/etc/vservers [04:39] initrd-Directory: $(sysconfdir)/init.d [04:39] pkgstate-Directory: ${prefix}/var/run/vservers [04:40] so you should probably do 'touch /usr/local/etc/vservers/.defaults/nonamespace' [04:41] thats it [04:41] seems to work now [04:41] ty Enrico [04:42] ensc: do you still run 0.09 on your laptop? [04:42] yep [04:42] did you ever try the chbind? [04:43] BTW Bertl...just tried a netstat and I can confirm that the vserver sees details of the host [04:44] Bertl: looks good; why do you ask? [04:44] read the line above ... [04:44] what does ifconfig give you after chbind? [04:44] (both) [04:45] it tells the host address [04:45] and in 1.26 too [04:46] hmm, do you ahve only one ip? [04:47] on vs1.26 I can't see IPs (addr) of the host ... [04:47] within a ctx, the ip is not displayed [04:47] with 0.09++ I see everything, and netstat reports all connections ... [04:48] # chcontext chbind --ip 1.2.3.4 ifconfig [04:48] New security context is 49154 [04:48] ipv4root is now 1.2.3.4 [04:48] eth0 Protokoll:Ethernet Hardware Adresse 00:D0:59:31:A2:83 [04:48] UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 [04:48] RX packets:1353 errors:0 dropped:0 overruns:0 frame:0 [04:48] this is with 0.09? [04:48] yep [04:49] jes, could you try that test too? [04:49] chcontext chbind --ip 1.2.3.4 ifconfig [04:49] hmmm just noticed when I start my vserver up it doesn't get the IP address [04:49] [root@cerberus skel]# chcontext chbind --ip 1.2.3.4 ifconfig [04:49] New security context is 49177 [04:49] ipv4root is now 1.2.3.4 [04:49] eth0 Link encap:Ethernet HWaddr 00:E0:81:51:DA:1F [04:49] UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 [04:49] RX packets:53260 errors:0 dropped:0 overruns:0 frame:0 [04:49] TX packets:50235 errors:0 dropped:0 overruns:0 carrier:0 [04:49] collisions:0 txqueuelen:1000 [04:49] RX bytes:15299199 (14.5 Mb) TX bytes:21040672 (20.0 Mb) [04:50] Interrupt:24 [04:50] lo Link encap:Local Loopback [04:50] UP LOOPBACK RUNNING MTU:16436 Metric:1 [04:50] okay ... [04:50] RX packets:8447 errors:0 dropped:0 overruns:0 frame:0 [04:50] TX packets:8447 errors:0 dropped:0 overruns:0 carrier:0 [04:50] collisions:0 txqueuelen:0 [04:50] RX bytes:1520439 (1.4 Mb) TX bytes:1520439 (1.4 Mb) [04:50] what about chbind --ip 1.2.3.4 ifconfig [04:50] does this report the same? [04:50] [root@cerberus skel]# chbind --ip 1.2.3.4 ifconfig [04:50] ipv4root is now 1.2.3.4 [04:50] eth0 Link encap:Ethernet HWaddr 00:E0:81:51:DA:1F [04:50] inet addr:192.168.1.100 Bcast:192.168.1.255 Mask:255.255.255.0 [04:50] UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 [04:50] RX packets:53308 errors:0 dropped:0 overruns:0 frame:0 [04:50] TX packets:50281 errors:0 dropped:0 overruns:0 carrier:0 [04:50] collisions:0 txqueuelen:1000 [04:50] RX bytes:15303756 (14.5 Mb) TX bytes:21048888 (20.0 Mb) [04:50] Interrupt:24 [04:51] lo Link encap:Local Loopback [04:51] inet addr:127.0.0.1 Mask:255.0.0.0 [04:51] UP LOOPBACK RUNNING MTU:16436 Metric:1 [04:51] and what enrico, and jes, does the following sequence: [04:51] RX packets:8447 errors:0 dropped:0 overruns:0 frame:0 [04:51] TX packets:8447 errors:0 dropped:0 overruns:0 carrier:0 [04:51] collisions:0 txqueuelen:0 [04:51] heh [04:51] RX bytes:1520439 (1.4 Mb) TX bytes:1520439 (1.4 Mb) [04:51] osnews.com really messed this story up [04:51] Solaris Express 02/04 is now available with exciting new features: partitioning a single Solaris instance into isolated application environments called "zones." It's a project similar to Linux Virtual Server (LVS). [04:51] hehe [04:51] lol [04:51] ifconfig eth0:1 192.168.0.1 [04:52] chbind --ip 192.168.0.1 ifconfig [04:52] [root@cerberus skel]# chbind --ip 192.168.0.1 ifconfig [04:52] ipv4root is now 192.168.0.1 [04:52] eth0 Link encap:Ethernet HWaddr 00:E0:81:51:DA:1F [04:52] inet addr:192.168.1.100 Bcast:192.168.1.255 Mask:255.255.255.0 [04:52] UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 [04:52] RX packets:53484 errors:0 dropped:0 overruns:0 frame:0 [04:52] TX packets:50438 errors:0 dropped:0 overruns:0 carrier:0 [04:52] collisions:0 txqueuelen:1000 [04:52] RX bytes:15320048 (14.6 Mb) TX bytes:21073632 (20.0 Mb) [04:52] Interrupt:24 [04:52] ah, see ... [04:53] eth0:1 Link encap:Ethernet HWaddr 00:E0:81:51:DA:1F [04:53] inet addr:192.168.0.1 Bcast:192.168.0.255 Mask:255.255.255.0 [04:53] UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 [04:53] Interrupt:24 [04:53] lo Link encap:Local Loopback [04:53] inet addr:127.0.0.1 Mask:255.0.0.0 [04:53] UP LOOPBACK RUNNING MTU:16436 Metric:1 [04:53] RX packets:8654 errors:0 dropped:0 overruns:0 frame:0 [04:53] TX packets:8654 errors:0 dropped:0 overruns:0 carrier:0 [04:53] collisions:0 txqueuelen:0 [04:53] RX bytes:1555578 (1.4 Mb) TX bytes:1555578 (1.4 Mb [04:53] errrr no ;) [04:53] 02:54 < jes> eth0 Link encap:Ethernet HWaddr 00:E0:81:51:DA:1F [04:53] 02:54 < jes> inet addr:192.168.1.100 Bcast:192.168.1.255 [04:53] that should not be visible ... [04:53] enrico, can you confirm? [04:53] Bertl: this happens with 1.26 too. Within a ctx, these info are hidden [04:54] ahhhh right [04:54] hum, you mean, it works with chcontext but not with chbind alone? [04:54] with chcontext, I do not see the IP of eth0 [04:54] without, I see it [04:55] interesting .. okay ... have to think about that ... [04:55] hmmm I'm lost too, my vserver isn't getting an ip address when I start it up at all [04:55] even though I've changed the .conf file [04:55] jes: there is no .conf file anymore... [04:56] ahhhh [04:56] that's considered a feature, not a bug ;) [04:56] do I set it within the vserver itself then? [04:56] interfaces are configured in a interfaces/ directory; just create a skeleton server with '--interface ...' option(s) [04:57] *nods* [04:57] (see 'vserver - build --help') [05:00] hmmm ok created the dev and ip files...still doesn't seem to read them though [05:00] http://vserver.13thfloor.at/Experimental/delta-2.6.3-vs0.09-vs0.09.5.diff [05:01] Is that for me Bertl? [05:01] okay, that should now have complete TB-sched support ... [05:01] whats TB sched? Or don't I want to know? [05:01] for whoever wants to try that ... [05:01] TB = Token Bucket ... [05:01] ahhhh errr ok ;) [05:01] basically that what the SILK slices do for planetlab ... [05:02] jes: what do you mean with 'seem'? How do you tested it? [05:02] "vserver skel start" [05:02] ifconfig will not show the create interfaces unless you give them a name [05:02] Topic changed on #vserver by Bertl!~herbert@MAIL.13thfloor.at: http://linux-vserver.org/ || latest stable 1.26, devel 1.3.7, exp 0.09.5 [05:02] use 'ip addr' instead of [05:03] ahhhh! [05:04] hmmm still something strange going on [05:04] okay, enough for me for today ... [05:05] have a nice wossname, everyone ... cu tomorrow ... [05:05] ahhh no it's ok [05:05] Nick change: Bertl -> Bertl_zZ [05:05] just me being silly [05:05] ok cya Bertl...ty for ALL the help ;) [05:06] so ensc, within the vserver I won't see the ipaddress with just a "ifconfig -a" ? [05:07] jes: you can make it visibly by creating a 'name' file in the interface's configuration directory [05:07] (name should contain the name) [05:07] whats the contents of that file? [05:08] like "eth0" ? [05:09] an arbitrary name, e.g. 'www' or 'ftp' [05:09] hmmm the file in - /usr/local/etc/vservers/skel2/interfaces/0 ? [05:10] yes, ...interfaces/0/name [05:11] ahhhh ok, works like a dream now...thanks Enrico [05:12] thanks for all your help again ensc...much appreciated [05:12] bedtime for me now....2:15am ;) [05:12] night all [05:12] jes (~jes@cpc1-leed5-3-0-cust196.ldst.cable.ntl.com) left irc: Quit: Leaving [05:16] youam (~youam@ciara.youam.de) joined #vserver. [05:19] virtuoso (~shisha@38ppp2.telegraph.spb.ru) left irc: Ping timeout: 480 seconds [06:07] Nick change: cdub -> cgone [06:42] Nick change: surriel -> riel [06:43] riel (~riel@riel.netop.oftc.net) left #vserver (Client exiting). [06:59] netrose (john877@FL3-24.217.241.239.charter-stl.com) left irc: Ping timeout: 480 seconds [08:12] nano_ (~nano@175.Red-217-127-186.pooles.rima-tde.net) joined #vserver. [08:14] nano_ (~nano@175.Red-217-127-186.pooles.rima-tde.net) left irc: Client Quit [09:15] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [09:15] ben- (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [09:18] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [09:18] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [09:24] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [09:25] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [09:37] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [09:46] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [09:53] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [10:00] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [10:06] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [10:10] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [10:22] IH (trilluser@h24-83-40-35.vc.shawcable.net) joined #vserver. [10:22] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [10:22] anyone home? [10:26] Hi IH [10:26] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [10:35] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [10:35] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [10:36] hi [10:36] i have a question about IP's in vservers [10:37] does each vserver in a physical server require a unique IP, or can they share a common IP designated to them? [10:44] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [10:44] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [10:44] they can share, but then you're fighting over ports... for example - you can't have two sshds running on port 22. [10:45] hmm, and I think you can get into some trouble with startup scripts [10:45] but i'm not sure about that [10:46] I run with few 'private' vservers running on internal network, no problem here. [10:46] Nope, no problem with startup scripts [10:48] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [10:48] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [10:54] ya port overlap would be a problem. [10:54] oh well i guess i need unique ips [11:01] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [11:02] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [11:06] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [11:08] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [11:11] Nick change: _maharaja -> maharaja [11:20] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: No route to host [11:27] ensc (~ircensc@ultra.csn.tu-chemnitz.de) left irc: Ping timeout: 480 seconds [11:30] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [11:31] ensc (~ircensc@ultra.csn.tu-chemnitz.de) joined #vserver. [11:32] hrm... [11:32] herbert? [11:32] thought you said you could change the hosntmae? :) [11:32] -bash-2.05b# hostname [11:33] test3.office.tektonic.net [11:33] -bash-2.05b# hostname test1 [11:33] hostname: you must be root to change the host name [11:39] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: No route to host [11:41] there's a cap for that i think [11:41] admin probably [11:41] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [11:46] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [11:46] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [11:52] ensc (~ircensc@ultra.csn.tu-chemnitz.de) left irc: Ping timeout: 480 seconds [11:59] eyck: yes, but dangerous [11:59] unless setting CAP_SYS_ADMIN isn't dangerous anymore? :) [12:00] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [12:03] kestrel (~athomas@syd-h43C.adsl.AlwaysONLINE.net.au) left irc: Quit: bbl [12:07] rs (rs@ice.aspic.com) joined #vserver. [12:07] hi [12:07] hello [12:13] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [12:17] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [12:17] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [12:19] stubbsd (~stubbsd@217.206.216.194) joined #vserver. [12:19] morning all, [12:19] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [12:23] [HvD] (~guess@62.99.252.14) got netsplit. [12:26] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [12:34] [HvD] (~guess@62.99.252.14) got lost in the net-split. [12:38] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [12:38] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [12:43] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [12:44] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [12:50] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [12:50] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [12:55] IH (trilluser@h24-83-40-35.vc.shawcable.net) left irc: Quit: http://isohunt.com [13:03] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [13:04] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [13:06] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [13:11] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [13:22] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [13:23] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [13:24] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [13:38] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [13:46] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Ping timeout: 480 seconds [13:50] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [14:01] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Ping timeout: 480 seconds [14:07] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [14:09] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [14:12] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [14:18] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [14:35] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [14:47] ensc (~ircensc@ultra.csn.tu-chemnitz.de) joined #vserver. [14:47] bengrimm (~ben@bengrimm-host225.dsl.visi.com) left irc: Ping timeout: 480 seconds [14:51] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [14:54] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [14:57] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [15:01] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [15:11] ensc (~ircensc@ultra.csn.tu-chemnitz.de) left irc: Ping timeout: 480 seconds [15:15] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [15:28] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [15:31] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [15:40] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Ping timeout: 480 seconds [15:46] paul (~irssi@muedsli-wan081.citykom.de) joined #vserver. [15:47] hi [15:52] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [15:57] bengrimm (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [15:58] bengrimm_ (~ben@bengrimm-host225.dsl.visi.com) left irc: Read error: Connection reset by peer [16:02] stubbsd (~stubbsd@217.206.216.194) left irc: Ping timeout: 480 seconds [16:07] virtuoso (~shisha@190ppp4.telegraph.spb.ru) joined #vserver. [16:31] virtuoso (~shisha@190ppp4.telegraph.spb.ru) left irc: Ping timeout: 480 seconds [16:43] broo (~broo@host30-5.btbx.net) joined #vserver. [16:43] stubbsd (~stubbsd@217.206.216.194) joined #vserver. [16:52] stubbsd (~stubbsd@217.206.216.194) left irc: Ping timeout: 480 seconds [16:59] stubbsd (~stubbsd@217.206.216.194) joined #vserver. [17:19] ensc (~ircensc@ultra.csn.tu-chemnitz.de) joined #vserver. [17:19] ensc (~ircensc@ultra.csn.tu-chemnitz.de) left irc: Client Quit [17:21] ensc (~ircensc@ultra.csn.tu-chemnitz.de) joined #vserver. [17:29] serving (~serving@213.186.190.121) left irc: Read error: Connection reset by peer [18:00] Nick change: Bertl_zZ -> Bertl [18:01] serving (~serving@213.186.190.121) joined #vserver. [18:02] hi everyone! [18:02] Hi Bertl! [18:05] Good day to you all. [18:05] =) [18:06] hello click, how are you? ;) [18:08] broo (~broo@host30-5.btbx.net) left #vserver (Client exiting). [18:13] Quite fine, just cleaning up some zonefiles, getting ready to transfer the secondary NS'es and MX's to another system. Low TTL when moving = good thing :) [18:26] (matta) hrm... [18:26] (matta) herbert? [18:26] (matta) thought you said you could change the hosntmae? :) [18:26] (matta) -bash-2.05b# hostname [18:26] (matta) test3.office.tektonic.net [18:26] (matta) -bash-2.05b# hostname test1 [18:27] (matta) hostname: you must be root to change the host name [18:27] hi matt! [18:27] yeah, you need CAP_SYS_ADMIN for now, but that will change in a few days ... [18:30] hi [18:30] is CAP_SYS_ADMIN ok to use? [18:30] looks like it allows a lot... [18:31] with CAP_SYS_ADMIN I could mount within a vserver [18:31] and like, it stayed inside the vserver [18:31] yep, that is why we are working on the vserver caps [18:31] don't know if that is intended... [18:31] actually the vserver caps is a long time project for no reason at all ... [18:32] everytime I satrt with it, something else comes along, which seems more important ;) [18:33] looks like a lot more people want the virtualized network than first though [18:33] thought [18:33] oh yeah [18:33] i got qemu working, seems faster than vmware [18:34] does it support multiple images? [18:34] so I can make another image as /vservers ? [18:35] yes [18:35] I use a minimalized version of the qemu setup, with no graphical frontend [18:35] this is more than sufficient for server setups [18:35] yeah [18:35] sec [18:35] how do I disable the frontend? I did it one time by accident... [18:36] qemu-0.5.2/i386/qemu-fast -nographic -L qemu-0.5.1/pc-bios -m 128 -snapshot -hda IMGs/TEST_32M.img -hdb IMGs/TEST_256M.img -kernel /usr/src/MORTON/linux-2.6.3-vs0.09.5-P1/arch/i386/boot/bzImage -append "rw root=/dev/hda1 max_vroot=1000" [18:36] this is an example line [18:36] you see how to specify memory, 2 disks and kernel args [18:36] the -snapshot feature ensures that the disk images are not modified [18:36] yeah, looks simple [18:37] the 32M image is my root filesystem, the 256M the vserver partition ;) [18:38] now i'll be able to test 2.6 [18:38] good! [18:38] ensc: are you around? [18:38] ensc (~ircensc@ultra.csn.tu-chemnitz.de) left irc: Ping timeout: 480 seconds [18:39] guess that was a no ;) [18:39] Bertl: interested in going a vserver patch for 2.6.3-wolk1.0? [18:39] s/going/doing [18:39] hmm, depends on the wolk1.0 ... [18:39] is it available in small pieces? [18:40] what is small pieces? :) [18:40] or just as a large bloob [18:40] it's a patch like the -mm patch [18:40] so not broken out patches? [18:40] s/not/no/ [18:40] I can upload them [18:40] mom [18:41] mom? [18:41] moment (just a moment) [18:41] one moment please == mom [18:41] :) [18:43] lol [18:43] ensc (~ircensc@ultra.csn.tu-chemnitz.de) joined #vserver. [18:52] ensc: you are here now? [18:53] Bertl: on its way. [18:54] ETA ~11mins [18:54] hum, that large? [18:54] 3,8mb bz2'ed [18:54] okay, is it platform agnostic? [18:54] 21mb unpacked, 103 files [18:55] agnostic? [18:55] does it work on all archs ... [18:56] dunno. I have x86 only [18:56] or does it break for example sparc [18:56] did anyone test the TBF scheduler? [18:56] it works like 2.6.3-mm4 works. maybe pax breaks some stuff, dunno about it [18:57] ExpiryJames (~james@h24-71-63-164.ok.shawcable.net) joined #vserver. [18:58] good morning [18:58] salve ExpiryJames [18:58] i guess not much to test besides "does it work" as the implementation itself has already been tested [18:59] you are optimistic today, right? [19:00] hehe [19:00] well the scheduler stuff was tested some time ago, the kernel scheduler might have changed, or I might haven broken it ;) [19:00] do not forget the last patch was on 2.4.x-ck, not 2.6.x [19:01] hi James! [19:02] did see your join, but missed the greeting, thought it was the 'typical' automatic join ... [19:03] Bertl: http://www.kernel.org/pub/linux/kernel/people/mcp/2.6-WOLK/broken-out/ [19:04] Bertl: wait some secs so you get the actual file list [19:04] master.ftp.kernel.org is somewhat slow ;) [19:05] maybe you come up with a vserver for wolk1.1 so I can integrate it :) [19:05] hhmm ... interesting ... [19:05] ficken.tar.bz2 27-Feb-2004 07:55 1.6M [19:05] lol [19:05] heh [19:05] it was a random name like I always choose for temporary tar's [19:06] :) [19:06] hotpr0n.patch [19:06] so that's it? [19:06] or should there be something else? [19:06] wait like i said [19:07] okay ... waiting for christmas ;) [19:09] hmm, files are gone? he? [19:11] ok now its there [19:11] :) [19:12] okay, mirrored it ;) [19:13] :) [19:14] http://vserver.13thfloor.at/Stuff/wolk.md5 [19:14] please verify ... [19:16] all ok [19:16] okay, I'll have a look at it ... this is wolk-1.0 ? [19:17] not 100%. it's what I have currently. it's somewhat in between 1.0 and 1.1 [19:17] not much has changed [19:17] hrm, okay, any internal numbering? [19:17] nope. I have to add some info's to each patch [19:17] so that you can merge something back later? [19:18] good that you said that [19:22] wolk1.0 is mainly up to 059 [19:22] so reiserfs and the tweaks are not included, right? [19:23] ah you have the mason patches, so quota for reserfs is included? [19:25] no. quota for 2.6 is not ready yet [19:28] ok, updating broken-out location with info in the top of each patch when which patch was added [19:48] okay, dinner time ... back in 20 [19:49] Nick change: Bertl -> Bertl_oO [20:04] matta (matta@tektonic.net) left irc: Quit: If at first you don't succeed, destroy all evidence that you tried. [20:09] monrad (~monrad@213083190235.sonofon.dk) joined #vserver. [20:10] Nick change: Bertl_oO -> Bertl [20:10] hi monrad! [20:10] hi Bertl [20:10] Bertl: its up again [20:11] one of my friends told me about vserver and now i "miratede" my main debian server to run with vserver it is much easier to administrate so far [20:12] sounds good ... [20:12] any troubles with that so far? [20:13] i had som problems setting up my mail because i am useing amavis to scan the mail but after i told it to use the right ip istead of localhost i work like a charm [20:14] mcp: http://vserver.13thfloor.at/Stuff/wolk.md5 please verify again ... [20:15] ok again [20:15] monrad: so you are here to enjoy the vserver community ... [20:16] Bertl: well yes i think irc is a nice way to pick up some tips to get more out of the software [20:16] Bertl: whos admining the mailinglist? [20:17] Martin [20:17] Bertl: i get more spam through that list than on any list I'm on [20:17] it ought to be fixed [20:17] just one question is it better to use the util-vserver package or the vserver one [20:17] click: well, obviously you do not read lkml [20:18] monrad: answer is simple: vserver tools are not maintained at the moment, and the util-vserver follow the changes in the kernel very closely ... [20:18] well, I dom but anything there are filtered on the server, and I hate adding secondary filters to ML's. [20:19] Martin is doing a lot of filtering, unfortunately even good filters are tricked by latest spams ... [20:19] and I definitely prefer an open ml over a closed one ... [20:19] well, try, but one of these were actually pretty clear-text [20:19] uh, typo [20:20] well, true, but one of these were actually pretty clear-text [20:20] IIRC martin was planning to provide the spam assasing header in the mails, haven't checked if he is already doing so ... [20:20] Nick change: cgone -> cdub [20:21] hi chris! [20:22] Bertl: I'll check, got a mail here [20:22] matta (matta@tektonic.net) joined #vserver. [20:22] mmmno. [20:23] just the additional antivirii header [20:23] no SA-support it seems, yet that is [20:24] just contact him, in a nice way, and ask him if he could verify why the spam assasin header isn't present ... [20:24] yo [20:24] what qemu bios do you use again? [20:24] click: that should help people to select the level themselves [20:24] matta: I used the qemu 5.1 bios, but guess 5.2 bios will do nicely ... [20:25] through the list? :) [20:25] BIOS panic at rombios.c, line 1558 [20:25] hrm [20:25] that's what comes with it [20:25] or to him via *-owner@ ? [20:25] :) [20:25] whatever you consider appropriate ;) [20:25] qemu-fast -L /usr/local/share/qemu -hda vserver-2.6.img -m 160 -nographic -append "root=/dev/hda1" [20:26] this is 5.2? [20:26] qemu-0.5.2/i386/qemu-fast -nographic -L qemu-0.5.1/pc-bios -m 128 -snapshot -hda IMGs/TEST_32M.img -hdb IMGs/TEST_256M.img -kernel /usr/src/RELEASE/STABLE/linux-2.4.21-vs1.00-P1/arch/i386/boot/bzImage -append "rw root=/dev/hda1 max_vroot=1000" [20:26] 0.5.2 [20:26] hrm [20:26] this was my line ... let me try with 5.2 bios ... 1sec [20:26] it works if I run qemu linux.img [20:27] linux.img being their default image [20:27] two questions: a) you did patch the kernel with the qemu aptch? [20:27] b) you used a reduced (non pci/etc) config? [20:28] qemu-0.5.2/i386/qemu-fast -nographic -L qemu-0.5.2/pc-bios -m 128 -snapshot -hda IMGs/TEST_32M.img -hdb IMGs/TEST_256M.img -kernel /usr/src/MORTON/linux-2.6.3-vs0.09.5-P1/arch/i386/boot/bzImage -append "rw root=/dev/hda1 max_vroot=1000" [20:28] qemu patch? :) [20:28] this works here nicely) [20:28] okay .. sec [20:28] http://vserver.13thfloor.at/Stuff/QEMU/ [20:28] pc-bios? [20:28] I just didn't insatll the qemu ;) [20:28] so I have to point it at the source directoy ... [20:29] how can i set the description for the vserver so i can se it when running vserver-stat ? [20:29] # Description: MD 8.2 Development Server Unit [20:29] as the first line of the config ... [20:30] ah 10-4 [20:30] stubbsd (~stubbsd@217.206.216.194) left irc: Quit: Leaving [20:31] rs (rs@ice.aspic.com) left irc: Quit: leaving [20:31] matta: if you use the qemu with the softmmu, you do not need to patch it, but it's a lot slower ... [20:33] how fast is qemu anyway? [20:33] faster than vmware but slower than uml ? [20:35] problem was the image [20:36] well, it's approaching native speed ... [20:36] probably after further optimization, a total factor of 2-3 is possible ... [20:36] the main difference to uml is, that it also works on a different arch ... [20:37] but I won't use it as a replacement for UML based servers [20:37] why not? [20:37] you have UML servers? :) [20:37] several reasons: [20:38] no I have not ;) [20:38] a) QEMU isn't perfect yet (so there might be bugs/errors/problems) [20:39] b) memory footprint of QEMU is horrible, compared to UML [20:39] c) uml uses native linux on a 'virtualized' hardware ... where QEMU, emulates all ahrdware [20:40] lol you don't like alex do you? :) [20:40] huh? [20:40] always so harsh between you and him [20:41] well, sometimes he manages to make me really angry, especially such replies like 'Linux-Vserver don't have it!' [20:41] I see at last time community only finding Herberts errors :-\ [20:41] lol [20:41] Doener_zZz (~doener@pD9588C58.dip.t-dialin.net) joined #vserver. [20:41] which needs some translation first ... [20:41] his features are nice and all, but sticking with RedHat... [20:42] actually the code is unmaintainable ... [20:42] and yes, i had much more problems with his code thans yours [20:43] so for the qemu kernel, no pci drivers? what about ide? [20:43] is supported ... otherwise you would not be able to boot ... [20:43] that's what I thought [20:43] so 'etc' mean ISA/EISA.. [20:43] ide0 at 0x1f0-0x1f7,0x3f6 on irq 14 [20:43] hda: max request size: 128KiB [20:43] hda: 65537 sectors (33 MB) w/256KiB Cache, CHS=65/16/63 [20:43] /dev/ide/host0/bus0/target0/lun0:<4>hda: set_multmode: status=0x41 { DriveReady Error } [20:43] hda: set_multmode: error=0x04 { DriveStatusError } [20:43] p1 [20:43] hdb: max request size: 128KiB [20:43] hdb: 524289 sectors (268 MB) w/256KiB Cache, CHS=520/16/63 [20:43] /dev/ide/host0/bus0/target1/lun0:<4>hdb: set_multmode: status=0x41 { DriveReady Error } [20:43] hdb: set_multmode: error=0x04 { DriveStatusError } [20:43] p1 [20:43] this is a bootup from the ide system ... [20:44] now you see what I mean with 'not perfect yet' [20:44] isa is somewhat supported, because it is required for the network card ... [20:45] so... [20:45] just no PCI support :) [20:45] ensc (~ircensc@ultra.csn.tu-chemnitz.de) left irc: Read error: No route to host [20:45] do i need a specific network driver? [20:45] yes, the ne2000 driver [20:45] sorry, my browser is all fucked up [20:45] i just rebooted [20:45] NE*000 ethercard probe at 0x300: 52 54 00 12 34 56 [20:45] eth0: NE2000 found at 0x300, using IRQ 9. [20:45] and now in konqueror half the websites I have like 20pt fonts [20:46] NE2000.. i prolly still have a box of them somewhere [20:47] uh [20:47] so did I need to fdisk this image ? [20:47] ensc (~ircensc@ultra.csn.tu-chemnitz.de) joined #vserver. [20:47] fine, but you have a box of virtual cards inside QEMU ;) [20:47] or will -hda file.img make it /dev/hda1 automatically ? [20:47] no you need to partition that image ... you need a script for that? [20:48] i can't just use fdisk ? [20:49] yes, but you need some tricks to format/fill those parts then ,,, [20:49] Doener (~doener@pD9E12EB7.dip.t-dialin.net) left irc: Ping timeout: 480 seconds [20:49] ok [20:49] yeah [20:50] i need to set the cylinders [20:50] 18:45 < Bertl> hdb: 524289 sectors (268 MB) w/256KiB Cache, CHS=520/16/63 [20:50] but you can use any sector/head combo ... [20:50] do your scripts do this? [20:51] no but they can mount/format such a partitioned image ... [20:51] ok [20:51] i'm game [20:52] fuck. [20:53] looks like somebody is playing with the linux-vserver page *sigh* [20:53] can you easily revert? [20:54] yea, no problem ... [20:54] i assume the wiki has version control [21:03] so, old status restored ... [21:14] kyencer (~kyencer@216.89.233.129) joined #vserver. [21:15] hi kyencer! [21:15] hey [21:15] i'm workin on getting those alpha's setup, should be done by 8 p.m. tonite [21:15] so he called the vserver project the KKK ? [21:17] what is a kkkk? [21:19] in america [21:19] there is the KKK [21:19] which is a rascist group [21:19] perhaps that is what they meant [21:20] hmm, guess not ... looks more like they started by holding the k key ... [21:26] but what do I know of portuguese statnic brains? 8-) [21:26] s/statnic/satanic/ [21:30] soor_ (~as@pD951ADE1.dip.t-dialin.net) joined #vserver. [21:31] hi soor_! [21:32] soor (~as@p5080BD8F.dip.t-dialin.net) left irc: Ping timeout: 480 seconds [21:37] monrad (~monrad@213083190235.sonofon.dk) left irc: Remote host closed the connection [21:43] Bertl: Were going to try to have 1 Mandrake-Alpha box and 1 Debian-Alpha box for ya by EOD [21:43] Bertl: What kernels would you like on these? [21:44] Bertl: We'll patch em up and everything [21:44] linux vserver kernels? [21:44] Bertl: Yes [21:44] okay, hmm, if it works with a 2.6 kernel, it should work with a 2.4 too, so I#d go for 2.6.3-vs0.09.5 [21:45] many thanks for helping out here ... [21:45] I really appreciate it ... and the community will do so as well ... [21:46] btw, we need some text for your entry in the Hall of Fame ... [21:46] Bertl: np [21:46] iprone (~iprone@216.89.233.129) joined #vserver. [21:46] Bertl: Email Richard for that kinda stuff [21:47] okay ... [21:48] jdike around? [21:49] Bertl: I did have some 0.09 functionality inquiries [21:49] let's hear ... [21:49] Bertl: Specifically regarding the usage of the memory limits [21:50] hmm, okay ... [21:50] guess this would be a good oportunity to discuss this on the channel, matt is also interested in the memory limit stuff ... [21:52] matta: right? [21:53] Bertl: would be great [21:53] hi [21:53] i thought we discussed [21:53] i have this discussion memorized now we've had it so many times :) [21:54] 3 "limits", minimum, rss, and vm [21:54] in general it should be base, power of 2, power of 4 [21:54] great, so we might be able to get some patches for testing ready ... [21:54] ie. 32MB/64M/256M [21:54] the basic idea, (as info for you kyle) [21:55] if memory limits are set, the kernel should fake the kernel structure (for /proc/meminfo, etc) [21:55] don't represent guaranteed, rss is represented as "ram" and vm as "swap" [21:56] is that we account RSS and VM, which is currently done, you can see that in /proc/virtual//limit [21:56] and handle the VM limit like memory + swap [21:56] and the RSS limit as available RAM [21:57] which isn't correct, but should be sufficient for the purpose [21:57] k [21:57] then we jsut enforce RSS, where it is about allocating and forking [21:57] which isn't correct either, but should work perfectly ... [21:57] yes, should work much much better than just VM limits [21:58] for the meminfo [21:58] yeah [21:58] that also :) [21:58] we fake the output as VM max = swap + ram [21:58] and RSS max = RAM [21:58] we toyed around with FreeVPS as far as limits on memory, but vservers were not liking it too well [21:58] then we give RSS current as ram used [21:59] and VM current - RSS max as swap used [21:59] that should give almost 'natural' linux kernel behaviour without much modifications to the current system [22:00] in a second stage, we ensure that RSS min affects the host swapper [22:01] in that way, that contexts below RSS min are not swapped out ... [22:04] and that VM min is actually available as swapspace (pure administrative value) [22:04] and one step further ... we can use RSS soft, and VM soft to control the swapout/process penalty for using that space ... [22:04] does this sound good? [22:06] yeah [22:06] right now we're using process limits to control our hogs [22:07] this mem limit is going to revolutionize everything for us, it's awesome, we really appreciate your dedication to this project [22:07] hope we can give back what we take away from it [22:07] you do not take anything away, you just receive ;) [22:07] :-) [22:08] and now you are giving too .. that's the great thing about it ... [22:08] Kevin told me he bit off some of the TB stuff too [22:09] I'll help out with memory stuff as much as I can [22:09] okay, you have a machine ready for some testing? [22:09] oh yeah [22:10] matta: do you have your fakemem patches at hand? [22:10] hrm [22:10] if emailing you can send em to kyencerATeapps.com [22:10] http://vps2.tektonic.net/patch-2.4.22-ctx17a-fakemem.diff [22:11] thanks ... [22:11] ah ok, great [22:11] kyle don#t use them yet! [22:11] bertl: can you rediff for 0.09? [22:11] brb, grabbing some lunch. [22:12] ok [22:12] jes (~jes@cpc1-leed5-3-0-cust196.ldst.cable.ntl.com) joined #vserver. [22:12] Hi all [22:12] hi jes! [22:12] heya Bertl, how are you tonight? [22:13] good, thanks and you? [22:13] not bad thanks...looking forward to a relaxing weekend ;) [22:13] we are currently attacking memory limits ... [22:13] oh? [22:14] as in trying to limit them? Or trying to overcome them? [22:14] as in trying to make them perfect for vserver ... [22:14] so the former ... [22:14] ahhh [22:14] actually you just reminded me...quick question ;) [22:15] with the alpha tools, how do I start my server with a certain capability? [22:15] say like CAP_MKNOD temporarily so I can use "drakconf" to remove some software from within the vserver [22:16] echo "CAP_SYS_BOOT" >/etc/vservers/WWWW/capabilities [22:16] echo "CAP_SYS_ADMIN" >/etc/vservers/WWWW/capabilities [22:16] aha! obvious when you know how ;) [22:17] can you combine multiple caps on different lines? [22:17] jes: one cap per line [22:17] jes: yes [22:17] great ty [22:17] had to fiogure that out last night :) [22:17] lol matta [22:17] hmm, the knowledge on the channel increases from hour to hour ... [22:18] lol well mine certainly does...but that's because I started at zero ;) [22:18] I like that ;) [22:18] oh BTW Enrico...Bertl and I spotted the other day that the build method doesn't pull in the /lib64 directory on x86-64 [22:19] jes: which build-method? [22:19] skeleton? [22:19] no legacy [22:19] although presumably thats by design since the x86-64 port is relatively recent? [22:20] oh... legacy... this is broken in several aspects [22:20] ahh ok [22:21] well I built my first vserver using legacy, now I'm just trimming the fat out of it (hence the caps question) [22:21] linux-vserver is arch agnostic per design .. but some things, like the lib dirs are not well tested ... [22:23] netrose (john877@SP2-24.207.231.2.charter-stl.com) joined #vserver. [22:23] wow....I have drakconf running across my network using a vserver and Exceed....I love this vserver ;) [22:24] matta: did you read the translation, guess it says all ;) [22:26] would it be considered a *bad* idea to have vservers use NIS clients from the host? [22:26] a host NIS server I mean [22:27] hmm, IMHO NIS is broken ... [22:27] but yeah, it should work this way ... [22:27] broken with regard to vserver? Or just broken generally? [22:27] generally ... NIS+ did some attemts to fix this ... [22:27] ahhh ok [22:28] netinfo would have benn cool, but never reached a useable state [22:28] just wondering if it'd be nice to use from a uid/passwd maintenance point of view [22:28] probably better to use ldap [22:28] hmmm don't have much experience of ldap, time to dig the docs out ;) [22:30] ooo sweet...drakconf works fine from within a vserver to remove software from it [22:30] you guys really *have* thought of everything ;) [22:30] didn't think of that ... 8-) [22:31] lol maybe not...but it works...and that's the main thing [22:31] well, maybe a small advantage is that I use mandrake as my prefered distro ;) [22:31] lol [22:32] do disk quotas work with alpha tools? [22:32] yes, they are orthogonal to them atm [22:32] you need separate tools to configure the hashes and the limits [22:33] ahhh ok, I saw the stable ones the other day, just didn't know if they'd work with alpha [22:34] problem is more, 2.6 kernel doesn't have the quota patches ydet [22:34] ahhh ok [22:35] right food time for me now, catch ya later [22:35] ty again [22:35] cya [22:35] jes (~jes@cpc1-leed5-3-0-cust196.ldst.cable.ntl.com) left irc: Quit: Leaving [22:36] monrad (~monrad@213083190235.sonofon.dk) joined #vserver. [23:01] monrad (~monrad@213083190235.sonofon.dk) left irc: Remote host closed the connection [23:05] monrad (~monrad@213083190235.sonofon.dk) joined #vserver. [23:24] ScoT (scot@host217-43-254-46.range217-43.btcentralplus.com) joined #vserver. [23:25] ScoT (scot@host217-43-254-46.range217-43.btcentralplus.com) left #vserver. [23:28] ok, 0.29.197 is out; it allows manual usage of the new namespace stuff ('vnamespace --enter vserver ... enter'). Integration will happen when create/migrate stuff is ready. [23:29] sounds good ... [23:30] Nick change: Doener_zZz -> Doener [23:30] hmm... i like the hacker page :) [23:31] hehe [23:31] those guys really have to be pretty l33t *lol* [23:41] Bertl: I'm back [23:41] good ... [23:41] Bertl: Do you want me to proceed with the patches that matt provided? [23:41] nope, that won't work, but I'll adapt that in a few minutes ... [23:50] they will be very very old [23:52] _shur1 (~shushushu@vserver.electronicbox.net) left irc: Read error: Connection reset by peer [23:53] _shur1 (~shushushu@vserver.electronicbox.net) joined #vserver. [00:00] --- Sat Feb 28 2004