[00:41] bmoney (g0d@CC4-24.207.131.200.charter-stl.com) joined #vserver. [00:50] hello? [00:55] sladen: i used the release 1.22 to patch my kernel and i used the util-vserver-0.26.tar.bz2 version of the utils [00:55] is that wrong? [00:56] so this isn't compatible with RH9? [01:30] ensc (~ircensc@ultra.csn.tu-chemnitz.de) left irc: Ping timeout: 512 seconds [01:44] netrose (~john877@CC3-24.171.21.47.charter-stl.com) joined #vserver. [01:47] bmoney (g0d@CC4-24.207.131.200.charter-stl.com) left irc: [02:19] Nick change: Bertl_oO -> Bertl_zZ [03:09] kungfuftr (~scott@jadis.narnia.org.uk) got netsplit. [03:09] CosmicRay (~jgoerzen@glockenspiel.complete.org) got netsplit. [03:10] CosmicRay (~jgoerzen@glockenspiel.complete.org) returned to #vserver. [03:14] kungfuftr (~scott@jadis.narnia.org.uk) returned to #vserver. [03:33] phil_: if you fetch 0.29 it should work [04:07] okay [04:07] phil_ (~phil@20.22.204.212.sr2.DTM1.ip.versanet.de) left irc: Quit: Client Exiting [04:27] Nick change: Doener -> doener_aw [07:48] Nick change: doener_aw -> doener_zz [07:58] RAJ_Onlin (RAJ_Online@202.9.130.197) joined #vserver. [08:31] RAJ_Onlin (RAJ_Online@202.9.130.197) left irc: Ping timeout: 480 seconds [09:03] riel (~riel@riel.netop.oftc.net) left irc: Ping timeout: 480 seconds [10:00] loger joined #vserver. [10:32] noel- (~noel@pD9E0963D.dip.t-dialin.net) joined #vserver. [10:40] noel (~noel@pD9FFA7E6.dip.t-dialin.net) left irc: Ping timeout: 504 seconds [12:16] Nick change: noel- -> noel [13:20] serving (~serving@213.186.190.138) left irc: Ping timeout: 480 seconds [13:34] re [13:49] tanjix (~tanjix@p5091F054.dip.t-dialin.net) joined #vserver. [13:49] hi @ll [13:50] tanjix: hi [13:53] tanjix (~tanjix@p5091F054.dip.t-dialin.net) left irc: Client Quit [14:59] tanjix (ViRu_@pD904A061.dip.t-dialin.net) joined #vserver. [15:12] serving (~serving@213.186.189.219) joined #vserver. [15:34] phil_ (~phil@20.22.204.212.sr2.DTM1.ip.versanet.de) joined #vserver. [15:36] hi [15:39] hi [15:43] phil_ (~phil@20.22.204.212.sr2.DTM1.ip.versanet.de) left irc: Remote host closed the connection [15:44] phil_ (~phil@20.22.204.212.sr2.DTM1.ip.versanet.de) joined #vserver. [16:02] Nick change: Bertl_zZ -> Bertl [16:02] hi everyone! [16:03] hi bertl [16:04] how are you? [16:04] fine, thanks [16:04] all configuration issues solved? [16:04] still now [16:04] not [16:11] hmm, what issues remain? [16:16] that not all vservers start on starting up the hostsystem [16:24] Doener_zZz (~doener@pD9588786.dip.t-dialin.net) joined #vserver. [16:25] tanjix: which ones do not start up, what is the pattern? [16:32] doener_zz (~doener@pD9E12B01.dip.t-dialin.net) left irc: Ping timeout: 512 seconds [17:04] sry, was afk... you need the context number (S_CONTEXT) ? [17:10] infowolfe (~infowolfe@pcp04891550pcs.frnkmd01.md.comcast.net) joined #vserver. [17:11] hey Bertl, good morning [17:11] morning infowolfe! [17:12] tanjix: no I would like to know what the not-starting servers have in common ... [17:13] Bertl: can't follow you, sry [17:13] oh, bertl, have you seen ensc today? [17:14] enrico, are you around? [17:26] virtuoso (~shisha@ip114-115.adsl.wplus.ru) left irc: Ping timeout: 480 seconds [17:27] infowolfe: seems hes not there ... still without a network connection over the holidays ... [17:28] ah.. [17:31] what do you need, maybe I can help? [17:32] from the vskel passwd.pl i get the following: [17:32] -- /usr/sbin/vserver --nodev --silent cstrike1 exec /usr/bin/passwd --stdin root [17:33] /usr/lib/util-vserver/save_s_context: fork: Resource temporarily unavailable [17:33] hmm, looks like the process limit is too low ... [17:34] how do you suggest i fix it? [17:34] did you try his newest tools? [17:34] http://www-user.tu-chemnitz.de/~ensc/util-vserver/util-vserver-0.27.tar.bz2 [17:35] i was actually just about to :-p [17:35] involved in a furious discussion on freenode's #gentoo-dev about fc :-D [17:35] fc? [17:36] fc like fc1? [17:36] fibre channel [17:36] like fc-sf [17:36] since fc-al is "too slow" lmfao [17:36] ... at least according to some of the guys in #gentoo-dev [17:37] 01[09:36] ciaranm: The rig I saw in '00 was an ia32 NUMA box with 16 nodes, one switch for each cpu cabinet (4 nodes), 4 adapters for each node, one bridge for each disk cabinet. Or I might be 01confusing the bridges and switches. [17:37] stuff that makes me drool [17:38] virtuoso (~shisha@ip114-115.adsl.wplus.ru) joined #vserver. [17:39] hi virtuoso! [17:39] infowolfe: you are pasting black text again ... [17:39] sorry [17:39] nvm then :-p [17:49] it did it again [17:50] /usr/lib/util-vserver/save_s_context: fork: Resource temporarily unavailable [17:50] Bertl, anything you'd like to see? [17:57] hmm, yes, the vserver config ... [17:58] which particular part? [17:58] or email the whole works to you [17:58] web page? [17:58] ok [17:59] http://thor.hardcore-linux.net/cstrike1.conf [18:00] try to raise the -u 200 to -u 1000 just for a test ... [18:00] which kernel patches do you use on that host? [18:01] 2.4.23-vs1.3.1 [18:01] same behavior with -u 1000 [18:05] hmm, could you try with vs1.22 .. maybe a bug ... [18:06] it'd require a reboot [18:06] but no problem [18:09] hey bertl? why is 1.22 1.22 instead of 1.2.2 [18:10] ahh, you are the first one to notice _and_ ask ... [18:10] it's just odd... and seems out of the normal naming convention (and if i hadn't had a problem booting my router this morning due to it being 1.22 instead of 1.2.2 it wouldn't have been fresh on my mind :-D) [18:10] well, first I thought about a numbering scheme like 1.00, 1.01, 2.00, 2.01, 2.02 ... [18:11] but then decided to use the kernel scheme (even/odd) [18:12] that doesn't make sense though... [18:12] was 1.20 your 20th iteration? [18:12] nevermind... i just caught on [18:12] but to simplify that I 'just' removed the last dot for the stable releases, with the idea, that stable and devel would be easy to distinguish this way ... [18:13] building 1.22 now [18:13] so 1.22 is actually 1.2.2 [18:13] I guess I get into troubles if I do more than 9 bugfix releases ;) [18:13] lol [18:23] it's a 1.3.1 bug [18:23] util-vserver-0.27, vs1.3.1 causes the error... vs1.22 does not [18:23] /usr/bin/passwd: invalid option -- - [18:23] usage: passwd [-f|-s] [name] [18:23] passwd [-x max] [-n min] [-w warn] [-i inact] name [18:23] passwd {-l|-u|-d|-S|-e} name [18:26] doesn't help that gentoo's passwd won't accept passwords from stdin :( [18:26] actually, that doesn't seem right... [18:32] Bertl, that problem above seems to be a 1.3.1 bug... but i'm not sure where's it's manifesting itself... or why [18:32] okay ... [18:32] try to lower the -u to 200 again ... and then maybe to 100 [18:33] and just for the fun of doing it, reboot with 1.3.1 and try if it works after the reboot, but maybe? fails later on? [18:33] vs1.22 does the right thing... *but* i'm not sure if --stdin is supposed to be passed to /usr/bin/passwd OR to /usr/sbin/vserver [18:34] I guess not ... passwd clears ever input ... for security, maybe usermod would be a better choice? [18:34] Bertl, not my problem, i'm just going to replace the passwd.pl with a print statement :-D [18:34] hmm ... [18:34] after i do the testing for you [18:34] -u 200 and then -u 100? [18:35] yes ... and see if 1.3.1 works after a reboot ... [18:35] Nick change: surriel -> riel [18:38] -u 200 works [18:39] hi rik, how was your xmas? [18:41] Bertl -u 100 also works with vs1.22 [18:41] want me to go back to 1.3.1 now? [18:42] if it isn't a too big annoyance, yes please ;) [18:42] no problem [18:43] doing it now [18:46] checking 1.3.1 with -u 100 again [18:49] Bertl, it's 1.3.1 [18:49] /usr/lib/util-vserver/save_s_context: fork: Resource temporarily unavailable [18:49] okay, thanks ... I'll have a look at the code right now ... [18:50] let me know if you'd like me to test [18:50] i'll switch my /usr/src/linux back over to the 1.3.1 tree to make a diff easier for you [18:50] hmm, you could give the 'unofficial' 1.3.2 a try ... maybe I fixed it already ;) [18:51] umm, ok [18:51] Action: infowolfe rushes back to reverse the 1.3.1 patch [18:51] http://vserver.13thfloor.at/Experimental/patch-2.4.23-vs1.3.2.diff [18:52] a tip, regarding different kernel builds ... [18:52] you can do the following, without any issues: [18:52] a) tar xjf linux-2.4.23.tar.bz2 [18:52] b) cp -la linux-2.4.23 linux-2.4.23-vs1.22 [18:52] b) cp -la linux-2.4.23 linux-2.4.23-vs1.3.1 [18:52] b) cp -la linux-2.4.23 linux-2.4.23-vs1.3.2 [18:53] c) cd linux-2.4.23-vs1.22 [18:53] building vs1.3.2 [18:53] d) bzcat ../patch-2.4.23-vs1.22.diff.bz2 | patch -p1 [18:53] Bertl, i like patch -Rp1 though :-p [18:53] yeah sure ... whatever you prefer ;) [18:53] I like /usr/src/linux to actually be whatever i'm currently running normally [18:54] it just saves a _lot_ of time and doesn require much space, and ln -s linux-2.4.23-vs1.22 linux is an option too ;) [19:01] 1.3.2 reboot now [19:05] trying now [19:07] Bertl, it's present in 1.3.3 also [19:07] oops [19:07] s/1.3.3/1.3.2/ [19:07] /usr/lib/util-vserver/save_s_context: fork: Resource temporarily unavailable [19:09] Bertl, i'm thinking that it's directly related to the --nodev code portions [19:10] or perhaps the --silent portion [19:10] /usr/sbin/vserver --nodev --silent cstrike1 exec /usr/bin/passwd is what is being run [19:27] Bertl, any ideas? [19:27] not yet ... [19:28] alright, the test string is right there... it should be easy to see if the bug is still there (finding out what else broke is the pain :-p) [19:29] oh, and btw, 1.3.2 isn't ready yet... it hung starting sshd for about 3 minutes before i killed the process via ctrl-c and then rebooted back into 1.22 [19:32] Nick change: tanjix -> tanjix[afk] [19:32] Bertl, is there a guide for understanding what showattr outputs? [19:32] [root$thor]::cstrike1> showattr bin/mount [19:32] bin/mount 00008010 [19:34] what the hell... i can't modify any files "created" by vskel! [19:34] can't replace, can't modify, can't do anything to them, can't even delete them! [19:43] Action: infowolfe feels stupid [19:43] nevermind :-p [20:07] phil__ (~phil@49.18.204.212.sr1.DTM1.ip.versanet.de) joined #vserver. [20:07] phil_ (~phil@20.22.204.212.sr2.DTM1.ip.versanet.de) left irc: Read error: No route to host [20:10] stubbsd (~stubbsd@217.206.216.194) joined #vserver. [20:10] anybody around that could clear up a question for a newbe to vserver? [20:12] stubbsd, go for it [20:12] if it's easy, i can handle it, if not, Bertl isn't away, so he can take care of it :-D [20:13] thanks, I have setup a vserver but thing that try to bind to 0.0.0.0 fail inside it, [20:13] stubbsd, that's a security feature :-p [20:14] do I have to use the chbind inside the vserver? [20:14] stubbsd, no, you need to set things like apache "Listen *" to "Listen :80" [20:14] If the vserver is limited to only one IP, can I set it up so if a service binds to 0.0.0.0 that get that IP? [20:15] no [20:15] you have to specify that ip for that service afaik [20:15] sh**t.. [20:15] sorry, man... [20:15] I have a service that does allow me to :-( [20:15] what service? [20:15] krb524d [20:15] Bertl: ping [20:17] stubbsd, what distribution is the vserver? [20:18] I've cooked the kernel from the 1.22 [20:18] and the 2.4.23 kernel [20:18] stubbsd, what i'm asking, the virtual server you're running, is it redhat, mandrake, suse, other? [20:18] gentoo. [20:19] :-D [20:19] sorry. [20:19] Action: infowolfe hugs stubbsd [20:19] i run gentoo :-p [20:19] cool. [20:19] how virtualized are you? [20:20] i have gentoo running inside my vservers and on my host with *no* startup problems :-D [20:22] sorry about that, the gnome "go have a brake came up, just killed it." :-) [20:22] lol [20:22] infowolfe, how did you do it, what about the depend's in the init scripts? [20:22] :-D [20:22] i'm +v on #gentoo-dev on freenode :-p [20:23] here's the basics... [20:23] keep getting kicked from gentoo-dev :-( "I'm just a low life.." [20:24] lol [20:25] i made friends with some devs in #gentoo and had some questions on building ebuilds [20:26] lol [20:26] umm, what're your current flood settings for /m? [20:26] confuison strikes. [20:26] ? [20:26] actually, nevermind [20:27] infowolfe, what are flood setting, [20:27] http://thor.hardcore-linux.net/fixme.sh [20:28] umm, they're settings for your irc client that automatically ignore someone if they flood you with too many messages in too short of a time period [20:28] cool thanks, [20:28] that link is where to get the "fixme.sh" [20:28] you should run it inside the vserver from / [20:28] got it, simple but great. [20:29] that's the first thing [20:29] what logger are you using in the vertual servers? [20:29] i've also edited most of the stuff in /etc/init.d/ [20:29] i use syslog-ng [20:29] my fails opening the kernel. [20:29] check http://thor.hardcore-linux.net/syslog-ng.conf for the configuration file i use [20:30] star, [20:31] that goes in /etc/syslog-ng/ [20:31] then just restart it and you're good [20:31] and you should *really [20:31] back in a min. :-) [20:32] *really* add /sbin/depscan.sh as the 11th line of /sbin/rc (it's a plain text file, just edit it with your favorite editor) [20:40] phil__ (~phil@49.18.204.212.sr1.DTM1.ip.versanet.de) left irc: Quit: Client Exiting [20:45] infowolfe, just done that and on startup depscan.sh fire off **lots** of awk proccess (just brougth the machine down) :-( [20:46] that's 2G of RAM, lol [20:46] sorry, not dead, just pooly [20:47] stubbsd, then something else is wrong [20:48] stubbsd, how did you build the vserver? [20:49] I thing I put the depscan.sh in the wrong file.... (lol) sorry one min... [20:49] where did you put it? [20:52] stubbsd, depending on how fast your connection is, i can tar a "fixed" gentoo vserver for you [20:52] /etc/rc.conf --------- it ran, and ran, :-) [20:52] if you don't mind apache, postfix, mysql, mod_php and some other stuff [20:53] oops :-p [20:53] lol [20:53] umm, /etc/rc.conf != /sbin/rc :-p [20:53] but i'm sure you know that now [20:54] btw, depscan.sh regenerates all your runtime directories and locks'n'stuff for starting a gentoo distro :-D [20:55] what else did you do to the init.d files? [20:55] thank, [20:56] s/thank/thanks/ [20:56] do you mind having apache, postfix, syslog-ng, mysql, mod_php, gentoolkit, and a few other toys installed by default? [20:56] I could easly clean them up, so no. [20:57] ok [20:57] then i'll start the tar [20:57] infowolfe, I owe you a pint! [20:57] no you don't [20:57] you owe me your soul, bwa ha ha ha ha (just kidding) [20:58] Action: stubbsd runs like hell [20:58] can I get a piece of it? [20:58] sure... but only if you use it for collision testing :-p [20:58] you know what they say, in for a peny in for a pound. [20:58] lol [20:59] infowolfe, where are you based? [20:59] baltimore, md [20:59] stubbsd, what kinda internet connection do you have? [20:59] 2mbit, [20:59] hdsl [20:59] hrm... (does some math in head) [20:59] it'll take you about as long to download this as it'll take you to download a redhat iso [21:00] and it's latest and greatest gentoo 2004.0 [21:00] (should be up-to-date) including emerge sync unless they've committed a lot to cvs lately :-p [21:00] is that unstable ie ACCEPT_KEYWORDS=">x86" or something else? [21:01] no, ACCEPT_KEYWORDS="x86" not ~x86 [21:01] s/>x86/~x86/ [21:01] umm, ooh, what're your cflags? [21:01] great, [21:02] Action: stubbsd runs round the room, gime, gime gimeeeee... [21:02] http://thor.hardcore-linux.net/make.conf [21:02] that's what the minor stuff has been built with [21:02] (apache, mod_php, mysql, postfix) [21:03] are you using distcc? [21:03] not actively [21:03] are, its going on p4 not athlon-xp :-( [21:04] don't worry about it [21:04] it doesn't matter that much [21:04] does it still work [21:04] change the make.conf flags [21:04] and btw, apache doesn't start by default :-p [21:04] no worrys. [21:05] do you have a working mailer on your system? [21:05] not as such. [21:05] okay, I found the nproc issue! [21:05] it's nrpoc? [21:06] sweet [21:06] Action: infowolfe mutters about sydlexia [21:06] dangit... :( [21:06] lol [21:06] that wasn't on purpose [21:06] although quite funny :-D [21:06] Action: infowolfe laughs [21:06] yes, it's the nproc option, in combination with my beginning change to the rlimit system ;) [21:06] ah... [21:07] I guess I have to fix that ... before 1.3.2 is released ... [21:07] stubbsd, would you mind sending me the output of your cat /proc/cpuinfo? [21:07] no [21:07] stubbsd, chip@capalon.com [21:08] i'd like to compare it to see if there might be any runtime issues [21:08] most of the time, athlon-xp flags only really matter when doing multimedia afaik [21:12] hey stubbsd, i'm thinking it'd be a good idea to rebuild everything once you've got my image and then tar it back up for backup purposes [21:14] stubbsd: PING [21:15] email is on its way [21:15] stubbsd, i'm going to post a link to this thing, when you've got it, chroot into it (don't turn it on as a vserver yet) adjust the make.conf flags to your liking and then emerge -fe world && emerge -e world [21:16] is there any whay to say emerge world, rebuild everying, not just update? [21:16] thanks. [21:16] that's it [21:16] ok. [21:17] oh, and you'll definitely want to adjust make.conf to your own liking, it's currently setup the way *i* like it... [21:17] ready, [21:17] which can be seen @ http://thor.hardcore-linux.net/make.conf [21:17] bzip2'ing to save you some bandwidth :-p [21:17] what are you connected to? [21:17] huh? [21:18] 100Mbit Level3 [21:18] sorry, what size is you pipe, if you know what I me. [21:18] where are you located? [21:18] uk [21:18] traceroute thor.hardcore-linux.net [21:18] and let me know what the ping between you and me is [21:18] Action: stubbsd sulks in the corner, every onces pipe is bigger than his. [21:19] stubbsd, don't fret... i know a gentoo dev that is on dialup! [21:19] hehe [21:19] happy new year everyone :) [21:19] happy new year... [21:21] stubbsd, nice box, btw [21:21] Action: infowolfe thinks everyone has a faster processor than him... [21:21] not want to play fair no pinging thor.hardcore-linux.net [21:22] it only takes about 40 mins for an iso anyway. [21:22] my own box is a apple 17" powerbook running gentoo (ppc) :-) [21:23] stubbsd, someone on your end must be filtering icmp, because i don't :-p [21:23] oh well, [21:24] infowolfe, how's the baking going. [21:25] stubbsd, go looking in about 15 minutes for http://thor.hardcore-linux.net/gentoo.tar.bz2 (or when you get the email) [21:25] cool, thanks [21:25] i have to run out for a few minutes, i'll ping you when i'm back [21:25] I'm going to have to be off, wife. I will see you around. [21:27] squedualed for 30mins :-).. see you [21:28] stubbsd (~stubbsd@217.206.216.194) left #vserver (Leaving). [21:45] netrose (~john877@CC3-24.171.21.47.charter-stl.com) left irc: Ping timeout: 480 seconds [21:50] RoMo (rmoriz@rmoriz.cpan.de) joined #vserver. [21:50] hi [21:50] i have some problem with vserver 1.22 + vquota on 2.4.23 - software raid1 devices don't get synced on reboot [21:50] anyone knows this problem? [22:24] ok seems to be a 2.4 kernel issue unrelated to vserver [22:44] good to know [23:37] netrose (~john877@CC3-24.171.21.47.charter-stl.com) joined #vserver. [23:45] Nick change: Bertl -> Bertl_oO [23:58] tanjix[afk] (ViRu_@pD904A061.dip.t-dialin.net) left irc: [00:00] --- Thu Jan 1 2004