--- Log opened wto maj 04 00:00:00 2004 --- Day changed wto maj 04 2004 00:00:00< ensc> Bertl: no 00:00:05< Bertl> hi axu, late hours for you? 00:00:56< axu> bertl: jepp ;) today(night) i want to upgrade my 2.4.22-ctx17a 00:01:53< Bertl> hey, what about a brand new 2.6.6-rc3 then? 00:02:22< axu> hehe, no sorry :) its revolto ... my "production"node :) 00:02:41< axu> bertl: try it with sun1 or 2 or delial ;) 00:03:19< axu> i thought bout some 2.4.whatever 00:03:35< axu> im not that happy with 2.6kernels performance/stability 00:03:52< Bertl> hmm, really? 00:04:36< axu> jepp :) tight now here on my home workstation (dual celeron) all seems much slower since i upgraded (first feel was fast ... but later) 00:05:24< axu> also i got saome oopses on sparc and intel with 2.6.3,4... 00:05:30< axu> :) 00:05:41< axu> i feel happy with 2.4 :) 00:05:54< axu> for private stable use :) 00:05:59< Bertl> okay, so 2.4.26-vs1.27 it will be? 00:06:11< axu> sozuunds nice :) 00:06:30< axu> a little like yoda 00:07:00< Bertl> well, not very adventerous you are, with stable all right it will be ;) 00:09:27< axu> 00:07:09 (4.09 MB/s) - `linux-2.4.26.tar.bz2' saved [30772389/30772389 :) 00:09:51< axu> letz get the rest of the incredients 00:09:53< taxcollector> Bertl: Speaking of "stable", are there any particular types of feedback on 1.3.x you are waiting for before declaring 1.4? 00:11:48< Bertl> well, yes, for example some testing ... 00:12:19< chaosle> night all 00:12:23>> chaosle [~yvan@port-212-202-168-55.dynamic.qsc.de] has left #vserver [Leaving] 00:12:37< Bertl> thing is, if I'd released 1.4 yet, it would have been _without_ a working barrier and unification code ;) 00:12:48< taxcollector> :) 00:13:02< Bertl> but obviously nobody tested/cared except for you of course ;) 00:14:09< Bertl> si I'm not really sure if I should release 1.4 at all ... 00:14:21< Bertl> s/si/so/ 00:14:37< taxcollector> And just continue plugging on the 2.6.x patches? 00:14:47< Bertl> and I didn't get any feedback for 1.3.x yet ;) 00:15:11< Bertl> at least folks are using the 2.6 branch (or it seems so ;) 00:15:52< Bertl> maybe I should do an opinion poll on the ml ... we'll see 00:17:00< taxcollector> From my position, I was looking forward to another 2.4.x stable release. I've done a rough patch of the 1.3.x code tree to the Red Hat Enterprise Kernel and I was hoping that I could follow 1.4 until Red Hat started using 2.6 00:17:19< broo> anyone know the easy way to tell the kernel to switch schedulers? The default one on 2.6.6-rc3 seems to be disk biased and causes long pauses in my gui under heavy disk load (like 2.4 used to) 00:17:41>> hiaslboy [matthias@i-166.vc-graz.ac.at] has joined #vserver 00:17:58< Bertl> broo: what you want is the magic swappiness arg in sysfs 00:18:47< Bertl> taxcollector: well, 1.3.x should be pretty stable, it probably needs a complete check, to see if all is working as expected 00:19:34< taxcollector> OK 00:20:11< Bertl> basically with 1.3.8 it was labeled 'release candidate' 00:20:49< broo> magic swappiness arg? somehow I don't think google will elighten me on that one 00:21:00< Bertl> sec 00:22:37< Bertl> 2.6 with swapiness of 0% is same as 2.4.19 - I assume 2.4.19's VM is 00:22:37< Bertl> the same as 2.4.26 (given feature freeze). 00:22:55< Bertl> /proc/sys/vm/swappiness 00:23:44< broo> swappiness = 60 00:23:59< broo> thx, I'll find some docs on it 00:24:12< Bertl> well, you've got a knob, now play with it ;) 00:24:49< broo> hehe, I see if I can figure out what the default scheduler was for 2.6.3 cause that one rocked 00:25:30< Bertl> you can do that too, but the long stalls on non RT tasks are the result of buffers being reclaimed ... 00:25:47< Bertl> (not of the scheduler decisions ;) 00:30:18< axu> does it make sense to run isakmpd in a vserver ? 00:30:33>> serving [~serving@213.186.189.95] has joined #vserver 00:30:36< axu> aeh.. wron os :) pluto&co 00:30:53< broo> whats iskmpd? 00:30:58< broo> whats isakmpd? 00:31:12< axu> broo: ipsec stuff obsd 00:31:48< broo> ah is see its an ike key manager 00:31:58< broo> Bertl: I would say yes 00:32:01< axu> jepp :) 00:32:14< Bertl> well, don't know it, but why not ... 00:32:45< axu> im just getting the superfreeswanpatches applied and then give it a try :) 00:33:27< Bertl> a free superswan *ooh* 00:40:45< axu> also... how about coda ? :) 00:41:05< Bertl> good idea ... 00:43:18< Doener> java is no fun (today|at all)... i'll go to bed... g'night everyone! 00:43:31< Bertl> night! 00:44:38< broo> hmm swapiness to the jerkiness out of the mouse but still have apps that don't respond to clicks or window shades and such, ah well 01:36:40>> hiaslboy [matthias@i-166.vc-graz.ac.at] has quit [Ping timeout: 480 seconds] 01:42:36>> infowolfe [~infowolfe@pcp04891550pcs.frnkmd01.md.comcast.net] has quit [Remote host closed the connection] 01:46:30>> infowolfe [~infowolfe@pcp04891550pcs.frnkmd01.md.comcast.net] has joined #vserver 01:46:59< albeiro> infowolfe: hi 01:48:11>> Bertl changed the topic of #vserver to: http://linux-vserver.org/ | latest stable 1.27, devel 1.3.9, exp 1.9.0pre14 01:49:20< infowolfe> hi albeiro 01:49:32< Bertl> okay pre14 should fix the attrib stuff (including the barrier) 01:49:46< Bertl> I'm off to bed now .. enjoy! 01:49:54< taxcollector> night! 01:50:02< infowolfe> nite Bertl 01:50:13< Bertl> night everyone! 01:50:18>> Bertl is now known as Bertl_zZ 01:50:22< albeiro> sleep well 01:50:47< albeiro> infowolfe: simple question - what do you think about gentoo vserver ? 01:51:23< infowolfe> albeiro, i'm not sure what you're asking 01:52:33< infowolfe> you mean vserver ON gentoo or gentoo inside vserver? 01:52:52< albeiro> both of them, but for now gentoo inside vserver 01:53:09< infowolfe> ok, what do you want to know about it? 01:53:16< infowolfe> /etc/portage/package.mask is your friend :-p 01:53:30< infowolfe> and most users don't know how to bypass it 01:53:48< albeiro> well, afaik gentoo devs thought about some subproject for vserver 01:54:42< infowolfe> who? and what? 01:54:56< albeiro> who ? Method for example 01:55:06< albeiro> what ? well, subproject of hardened 01:55:14< infowolfe> oooh 01:55:29< infowolfe> you mean making the vserver-project a subproject of hardened for gentoo hosts 01:55:38< albeiro> yes 01:55:44< infowolfe> including the kernel patches in the hardened-sources ebuilds 01:55:51< infowolfe> so what do you want to know? 01:57:26< albeiro> 1. if it can be done without much pain 2. what i need to change in system 3. how does it work - i mean, if there are problems with it, etc 01:57:47< infowolfe> can you give me a few minutes? 01:58:00< albeiro> yes, sure 01:58:04< infowolfe> i need to hit the store before they do a shift change (so i can get some mt dew) 01:58:07< infowolfe> be back in a flash 01:58:41< albeiro> ok, let's move it to hardened as you have started talking with method, ok ? 02:05:50< infowolfe> i'm back 02:05:57< albeiro> so go to -hardened 02:06:00< infowolfe> i really don't have the time to be full-time on this 02:10:36< axu> hmm, i upgradet vservers & tools, started the vservers but cant access them from the rootserver via vserver and vserver-stat, anyone a clue ? 02:11:08< axu> no big problem all services up and running and i can connect via ssh :) 02:14:00< infowolfe> axu, what kernel, patch, and util-vserver versions? 02:14:44< axu> # patch-2.4.26-vs1.27.diff + util-vserver-0.29.4.tar.bz2 02:15:29< axu> infowolfe, but, actually, lacy me hasnt even checked the configs for changes ;) 02:16:35< axu> i think i have a closer look at it when im awake :) 02:16:37< infowolfe> what's the output when you vserver enter 02:16:56< axu> vserver-stat: unable to switch in context security #1 02:17:39< infowolfe> try to kill the vserver 02:18:02< axu> :( but it running so nicely :) 02:19:14< infowolfe> i don't know what to tell you then 02:20:02< infowolfe> albeiro, what kv and pv are you running? 02:20:08< infowolfe> i need to get the latest exp 02:20:48< infowolfe> ensc, ping: what's the latest version of your tools? 02:21:09< albeiro> infowolfe: kv ? pv ? 02:21:24< infowolfe> kernel version and patch version 02:22:26< albeiro> nothing now, waiting till vserver gets into wolk, what will happen in few weeks 02:22:41< albeiro> just 2.6.5 kernel 02:23:20< infowolfe> vserver is getting into wolk sources? 02:23:53< taxcollector> infowolfe: 0.29.211 is the most recent alpha util-vserver (for 1.3.x and 1.9.x) on ensc's web site. 02:23:53< albeiro> is going to, after 2.6.6 kernel 02:26:24< infowolfe> taxcollector, do you have a link for the latest exp? 02:26:53< axu> thnkas infowolfe, ill check tomorrow :) 02:27:02< taxcollector> All the new alpha tools get placed at http://www-user.tu-chemnitz.de/~ensc/util-vserver/alpha/ 02:27:29< infowolfe> taxcollector, i was talking about a 2.6.5 kernel patch 02:28:16< taxcollector> Oh, OK -- all of Bertl's new experimental patches get placed at http://vserver.13thfloor.at/Experimental/ 02:28:21< taxcollector> pre14 is the latest one 02:28:32< taxcollector> but that is for 2.6.6 02:34:49>> lilo [levin@lilo.usercloak.oftc.net] has quit [Remote host closed the connection] 02:37:31< axu> bye folks, 02:37:34>> axu [gl@83.64.2.169] has left #vserver [Leaving] 02:41:13>> anonymouscoward [~nwalsh@shaggy.internode.com.au] has left #vserver [Kopete 0.8.1 : http://kopete.kde.org] 02:50:01< infowolfe> hey albeiro, check this out: http://c3.hardcore-linux.net/smp/ 02:50:14< infowolfe> that's my desktop 02:58:24< albeiro> i will :) 03:00:12< albeiro> nice 03:01:36< infowolfe> i <3 apache's mod_proxy :-p 03:02:06< infowolfe> i could probably do name based hosting from my desktop through my router if i wanted to 03:02:10< infowolfe> or something along those lines 03:02:47>> lilo [levin@lilo.usercloak.oftc.net] has joined #vserver 03:04:57>> infowolfe [~infowolfe@pcp04891550pcs.frnkmd01.md.comcast.net] has quit [Remote host closed the connection] 03:14:20>> ensc [~ircensc@ultra.csn.tu-chemnitz.de] has quit [Ping timeout: 480 seconds] 03:19:53>> infowolfe [~infowolfe@pcp04891550pcs.frnkmd01.md.comcast.net] has joined #vserver 03:19:54>> infowolfe_ [~infowolfe@pcp04891550pcs.frnkmd01.md.comcast.net] has joined #vserver 03:19:59>> infowolfe [~infowolfe@pcp04891550pcs.frnkmd01.md.comcast.net] has quit [Quit: ] 03:20:16>> infowolfe_ [~infowolfe@pcp04891550pcs.frnkmd01.md.comcast.net] has quit [Quit: ] 03:20:22>> infowolfe [~infowolfe@pcp04891550pcs.frnkmd01.md.comcast.net] has joined #vserver 03:20:30< infowolfe> ok 03:20:46< infowolfe> attempting to compile alsa with exp1.9.0pre14 is not a good thing to do :-p 03:24:13< albeiro> heh 03:25:50< infowolfe> i figure there were at least 1,000 or so EIP OOPSs before i could turn the box off 03:26:04< albeiro> rotfl 03:27:11>> taxcollector [~taxcollec@192.16.167.161] has quit [Quit: ] 03:36:58>> _id [~id@pD9519E32.dip.t-dialin.net] has quit [Ping timeout: 480 seconds] 03:44:19>> ensc [~ircensc@ultra.csn.tu-chemnitz.de] has joined #vserver 03:46:17>> _id [~id@pD9519E32.dip.t-dialin.net] has joined #vserver 03:51:11>> anonymouscoward [~nwalsh@shaggy.internode.com.au] has joined #vserver 04:46:24>> _id [~id@pD9519E32.dip.t-dialin.net] has quit [Ping timeout: 480 seconds] 07:50:50>> hiaslboy [matthias@i-166.vc-graz.ac.at] has joined #vserver 08:04:38>> CornerNote [~info@dsl-203-113-210-248.QLD.netspace.net.au] has joined #vserver 08:04:43< CornerNote> hello 08:04:56< CornerNote> i am looking for some help with SSH and vServer 08:05:11< infowolfe> CornerNote, how can i help? 08:05:34< CornerNote> my vserver is up and running.. but i cannot SSH directly into it 08:05:46< CornerNote> i can only get into it using the vserver enter command 08:06:18< CornerNote> when I ssh to the IP of the vserver, only the root (of the real server) will get me in 08:07:09< infowolfe> ok 08:07:17< infowolfe> what's your favorite editor? 08:07:26< CornerNote> vi 08:07:29< infowolfe> for i in `find /etc -name "sshd_config"`; do vi $i;done 08:07:51< infowolfe> you know the ip of the root server, right? 08:07:59< CornerNote> someone told me to change /etc/ssh/sshd_config 08:08:08< CornerNote> and change 0.0.0.0 to the servers IP address 08:08:22< CornerNote> ahhh - wait - i need to restart SSH which i didnt do 08:08:23< infowolfe> uncomment "#Port 22" and "ListenAddress 0.0.0.0" and change 0.0.0.0 to the host's ip address 08:08:27< infowolfe> lol 08:08:50< infowolfe> ok, so restart it, then reconnect to the host and from there, restart the sshd on the vserver and it'll work 08:10:08< CornerNote> it does not seem to be installed on the vserver 08:10:22< CornerNote> there is no /etc/rc.d/init.d/sshd 08:10:39< infowolfe> what distro is the vserver 08:11:03< CornerNote> hrmm.. i only installed it last week... let me check 08:12:09< CornerNote> vserver-0.29-1.i386.rpm 08:12:27< infowolfe> you misunderstand 08:12:39< infowolfe> cd to the vserver's "/" 08:12:48< CornerNote> yeah - i can do that sometimes :) 08:12:50< CornerNote> ok 08:12:51< infowolfe> ls /etc | grep release 08:13:08< CornerNote> redhat-release 08:13:21< infowolfe> so cat /etc/redhat-release 08:13:42< CornerNote> Red Hat Linux release 7.3 (Valhalla) 08:16:25< CornerNote> should i copy the sshd from the base servers /usr/sbin ? 08:18:44< infowolfe> no 08:18:50< infowolfe> where did you get the image? 08:29:59< CornerNote> off the RH7.3 CD 08:30:34< infowolfe> no 08:30:43< infowolfe> where did you get the vserver context image 08:31:30< CornerNote> yes - off the CD 08:31:40< CornerNote> when i installed the vserver it told me to insert the CD 08:31:53< infowolfe> well, that's past my realm of experience 08:32:05< CornerNote> then it went ahead and loaded RD7.3 onto the context 08:32:06< infowolfe> if the documentation can't help, neither can i (i'm not really a redhat guy) 08:32:21< CornerNote> ok - thanks for your help :) 08:32:30< CornerNote> i will try copying sshd from the main server 08:32:53< infowolfe> why don't you use the rpm? 08:33:07< CornerNote> ok :) 08:33:21< CornerNote> arent you supposed to use the special v_ssh ? 08:33:30< CornerNote> that comes with the vserver package 08:33:58< CornerNote> maybe thats why it's not on the vserver 08:43:33< CornerNote> do u use debian ? 08:43:42< infowolfe> no 08:43:43< infowolfe> gentoo 08:43:54< CornerNote> ahh - i heard good stuff about that 08:43:57< CornerNote> never used it 08:44:45< CornerNote> maybe i should try a different linux 08:45:30< infowolfe> or read the docs :-D 08:46:16< CornerNote> i have RTFM - over and over 08:46:24< infowolfe> ok 08:46:25< CornerNote> they are very underdeveloped 08:46:29< infowolfe> oh really? 08:46:41< CornerNote> i am reading this site 08:46:45< CornerNote> http://www.linux-vserver.org 08:49:08< infowolfe> ok 08:49:10< infowolfe> have you use google? 08:51:50< CornerNote> nope 08:51:57< CornerNote> i will try now 09:04:19< CornerNote> i am off home 09:04:27< CornerNote> thanks for your help infowolfe 09:04:30>> CornerNote [~info@dsl-203-113-210-248.QLD.netspace.net.au] has quit [Quit: ] 09:15:35>> _id [~id@pD95E9DF0.dip.t-dialin.net] has joined #vserver 10:13:08>> rs [rs@ice.aspic.com] has joined #vserver 10:13:20< rs> hi dudes 10:37:42< infowolfe> hi rs 11:08:07>> hiaslboy [matthias@i-166.vc-graz.ac.at] has quit [Quit: using sirc version 2.211+KSIRC/1.3.10] 11:10:38>> Doener` [~doener@pD9E12807.dip.t-dialin.net] has joined #vserver 11:16:46>> hiaslboy [matthias@e-16.vc-graz.ac.at] has joined #vserver 11:17:32< _id> r? 11:17:43>> Doener [~doener@pD95883FD.dip.t-dialin.net] has quit [Ping timeout: 480 seconds] 11:19:39>> hiaslboy [matthias@e-16.vc-graz.ac.at] has quit [Quit: ] 11:19:55>> hiaslboy [matthias@e-16.vc-graz.ac.at] has joined #vserver 11:27:11< Doener`> morning! 11:46:36>> no_maam [~erik@datenzone.de] has joined #vserver 11:46:38>> no_maam_ [~erik@datenzone.de] has quit [Read error: Connection reset by peer] 12:56:48>> Crux [~Crux@203.167.84.219] has joined #vserver 12:57:43>> Crux [~Crux@203.167.84.219] has left #vserver [Leaving] 12:58:56< Doener`> woohoo! debian vserver-patch has been updated from 1.21 to 1.27 :-D 13:03:13>> Bertl_zZ is now known as Bertl 13:03:22< Bertl> Doener`: great news! ;) 13:08:43< Bertl> rs: did you get around testing 1.9.0pre14 yet? 13:08:59< Bertl> (just curious) 13:27:13< Bertl> infowolfe: hmm, alsa doesn't like vserver? 13:30:50< Bertl> infowolfe: where is that phpSysInfo theme from? 13:38:15< rs> re 13:38:24< rs> now I am 13:38:43< Bertl> cogito ergo sum? 13:38:55< rs> Bertl: what was changed ? 13:39:16< Bertl> basically the attribute stuff was fixed ... 13:39:50< Doener`> Bertl: the theme is in the debian package of phpSysInfo, and as the demos on the phpSysInfo homepage also got this theme available i guess it's also in the tar.gz 13:39:58< rs> Bertl: kmsg ? 13:40:10< Bertl> no, barrier, iunlink + immutable ;) 13:40:36< Bertl> kmsg got a debug loggin message ... I'm currently investigating this issue ... 13:40:43< rs> was a pb with that ? 13:41:02< Bertl> yep, neither unification nor barrier where working ;) 13:41:03< rs> I did test unification yet 13:41:15< Bertl> s/where/were/ 13:41:16< rs> hehe :) 13:41:24< rs> I didn't test unification yet 13:41:36< Bertl> well, now you can .. ;) 13:42:22< rs> ok, but I have some other things to fix/do before, but I can test with just a couple of vservers 13:42:58< Bertl> np, I'm planning to release 1.9.0 this week, if everything seems okay ... 13:42:59< rs> about the memory logs I have, do you need it ? 13:43:24< rs> I have about 1.5Gb of logs 13:43:31< rs> uncompressed 13:43:39< Bertl> well, wont hurt to keep them around for some time, haven't got around analyzing the last one ... 13:44:04< Bertl> need to write some scripts to analyze it ... 13:44:05< rs> ok np 13:47:18< rs> no idea of where I can find info about how to deal with smp_affinity (cf you mail) ? 13:47:57< Bertl> well, its simple, it's a mask .. echo "3" >/proc/irq/5/smp_affinity 13:48:06< Bertl> will enable CPU 0 & 1 13:50:39< infowolfe> Bertl, it's stock with phpsysinfo 13:50:43< infowolfe> it's called "aq" 13:50:48< infowolfe> well, anyway it's stock with the gentoo install 13:51:37< Bertl> hmm, the sourceforge page has no demo of that, so I wasn't sure ... 13:51:54< rs> Bertl: ok but by default all mask are set to ff, so why should be setup a more restrictive one ? 13:52:21< Bertl> infowolfe: forget it, I'm too stupid for those interfaces .. found the popup list ;) 13:52:38< Bertl> rs: where are interrupts delivered atm? 13:52:40< infowolfe> and Bertl, when compiling alsa's drivers with my kernel 2.6 config, and exp1.9.0 it broke :- 13:53:02< Bertl> infowolfe: and without vs1.9.0 it was okay? 13:53:07< infowolfe> yup 13:53:20< Bertl> do you ahve any stack traces? (oopses) 13:53:31< infowolfe> they wouldn't stop long enough for me to trace 13:53:41< rs> Bertl: on cpu1 13:53:45< rs> CPU0 CPU1 13:53:45< rs> 0: 469684 80154600 IO-APIC-edge timer 13:53:48< infowolfe> like i said, there were probably about a thousand of them before i could get the machine to turn off 13:53:51< rs> for the interupt 0 13:54:06< Bertl> well, interrupt 0 isn't the best example ... 13:54:12< rs> but on cpu0 for the 16 13:54:15< Bertl> what about 13 or 14? 13:54:26< rs> they are not listed 13:54:32< Bertl> no ide? 13:54:40< rs> no :) 13:55:08< Bertl> hmm, okay, try to echo "3" to 0 and 2 13:55:20< Bertl> and save the current values somewhere ... 13:56:26< rs> echo "3" > /proc/irq/2/smp_affinity 13:56:29< rs> cat /proc/irq/2/smp_affinity 13:56:29< rs> ff 13:56:52< Bertl> that's okay ... did 0 change? 13:57:25< rs> CPU0 CPU1 13:57:25< rs> 0: 570283 80286430 IO-APIC-edge timer 13:57:25< rs> 2: 0 0 XT-PIC cascade 13:58:00< Bertl> 80286430-80154600 13:58:01< Bertl> 131830 13:58:01< Bertl> 570283-469684 13:58:01< Bertl> 100599 13:58:06< Bertl> looks better ;) 13:58:14< rs> yep :) 13:58:52< rs> I don't understand why but it's better ;o) 13:59:22< Bertl> seems like the default value of ~0 isn't the same as a specified mask ... don't ask me why, it's probably a bug ;) 13:59:43< Bertl> if you care about, you can post it on lkml ;) 14:01:31< rs> hum why not :) 14:22:07< Bertl> rs: hmm, the kmsg issue is strange ... 14:22:26< Bertl> # ls -la /proc/kmsg 14:22:34< Bertl> # chcontext --ctx 100 ls -la /proc/kmsg 14:22:34< Bertl> New security context is 100 14:22:34< Bertl> ls: /proc/kmsg: No such file or directory 14:42:45>> dsanta [~santa@68.190.156.105] has quit [Ping timeout: 480 seconds] 15:21:47>> Apollo [~throwaway@caracal.norcomcable.ca] has joined #vserver 15:27:08>> dsanta [~santa@c68.190.156.105.roc.mn.charter.com] has joined #vserver 15:28:00< Bertl> hi Apollo! hi dsanta! 15:44:31< Apollo> hi 15:44:43< Bertl> hmm, good response time ;) 15:44:57< Apollo> hehe... was thinking that after I hit enter. :) 15:45:01< Bertl> lilo usually needs 1-2 days ... 15:46:29< broo> vserver wiki won't let me regiser :( oh well, it still let me add the starting an X server in a vserver to the wiki faq 15:46:53< Bertl> broo: registering is cookies only 15:47:31< broo> it kept saying I didn't have a valid email address, but I've used it for many years now :) 15:47:48< Bertl> hum, are we talking about the same wiki? 15:48:05< broo> http://vserver.strahlungsfrei.de/tiki-register.php 15:48:14< Bertl> ak okay, not my business then ;) 15:48:24< broo> its the wiki pointed to off of linux-vserver.org 15:49:55< broo> ah I see the direct faq is a wiki style as well, should I have put it there instead? 15:50:05< broo> http://www.linux-vserver.org/index.php?page=Linux-Vserver+FAQ 15:50:12< Bertl> you can do that ... np 15:56:50>> rs [rs@ice.aspic.com] has quit [Ping timeout: 480 seconds] 16:16:37< broo> okay added it, hopefully people will understand it 16:17:26< Bertl> did you use the words: simple, easy, straight or small ? 16:18:06< broo> nope 16:18:17< Bertl> hmm, okay, then there is a small chance ;) 16:18:23< broo> I did use caveat 16:21:14>> rs [rs@ice.aspic.com] has joined #vserver 16:42:14< rs> re 16:57:22>> _id is now known as _id_m12 17:09:06>> Napalm123 [~napalm@host81-7-22-112.adsl.v21.co.uk] has joined #vserver 17:09:18< Napalm123> hello everyone 17:09:53>> Napalm123 is now known as Napalm 17:10:09< Napalm> is Herbert around? 17:10:34< Napalm> anyone? 17:10:52< broo> I believe he was here an hour ago, but don't know if he still is 17:11:29< Napalm> i only just got his email :-s 17:12:44< Napalm> well now im here, does anyone want to chat about the linux-vserver project 17:12:49>> shuri [~shushushu@cpu183.adsl.qc.bellglobal.com] has joined #vserver 17:13:58< broo> sure what about it do you want to chat about? :) 17:14:41< Napalm> well ive just set a test server up with it 17:14:55< Napalm> kernel 2.4.26-vs1.27 17:15:29< Napalm> i've just started with it, i've been reading up on the unification bits and bobs 17:15:34< Napalm> its a great idea --- Log opened wto maj 04 17:45:32 2004 17:45:32>> albeiro [albeiro@linux.gentoo.pl] has joined #vserver 17:45:32>> Irssi: #vserver: Total of 41 nicks [0 ops, 0 halfops, 0 voices, 41 normal] 17:45:40>> Irssi: Join to #vserver was synced in 8 secs 17:45:42< broo> looks like lisp code got stuck in your kernel code almost 17:46:47< Bertl> well, what compiler is that? 17:48:24< Nesh> 2.96 17:48:31< Nesh> Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/2.96/specs 17:48:31< Nesh> gcc version 2.96 20000731 (Red Hat Linux 7.3 2.96-113) 17:48:37< Nesh> trying to build 2.6.4 now. 17:49:06< Bertl> hmm, well aside from the fact that there is no 2.96 gcc, I'd try with 3.3.3 or 3.4 ... 17:49:11< broo> don't you need 3.x to build the 2.6 kernels 17:49:32< Bertl> broo: no, not necessarily ... 2.95.4 did compile 2.6.4 very nicely ... 17:49:38< Nesh> 3.x is a pain in the ass to get rpm's for rh7.3 17:49:46< broo> ah 17:49:57< Bertl> Nesh: you can use my cross compiling packages 17:50:06< Nesh> what/where that 17:50:09< Bertl> sec 17:50:32< Nesh> k 17:51:08< Bertl> i386, right? 17:51:11< Nesh> yes 17:54:10< Napalm> im back 17:54:18< Napalm> hello herbert 17:55:26< Nesh> hello Napalm 17:55:38< Napalm> hello sorry i dissapeared 17:55:42< Bertl> hi Napalm! stay away from the fire! 17:55:47< Napalm> lol 17:55:58< Napalm> i get the fix my computer phone calls 17:55:59< Napalm> ;) 17:56:52< Napalm> herbert shall i call you Bert 17:56:53< Bertl> Nesh: 17:56:55< Bertl> http://vserver.13thfloor.at/Stuff/Cross/binutils-i386-2.15.90.0.3-3mdk.src.rpm 17:57:03< Bertl> Napalm: if you like to ... 17:57:07< Nesh> Bertl k 17:57:29< Bertl> let's start with this baby ... you probably need to specify some args ... 17:57:54< Nesh> will this replace my binutils? 17:58:01< Bertl> no, you do the following: 17:58:04< Nesh> ok 17:58:11< Bertl> rpm -i 17:58:12< Napalm> btw bert the answer to your question is yes it will be opensource 17:58:13< Bertl> then 17:58:19< Bertl> Napalm: good! 17:59:20< Bertl> rpmbuild -ba /usr/src/redhat/SPEC/binutils-cross.spec --define "CROSS_ARCH i386" 18:00:22< Nesh> SPECS 18:00:23< Nesh> :P 18:00:24< Nesh> lol 18:00:35< Nesh> ok its building now 18:00:59< Bertl> can't get everything right, can I? 18:01:02< Nesh> creating Makefile 18:01:02< Nesh> + %make -j 5 18:01:02< Nesh> error: Bad exit status from /var/tmp/rpm-tmp.54042 (%build) 18:01:07< Nesh> error: Bad exit status from /var/tmp/rpm-tmp.54042 (%build) 18:01:16< Nesh> /var/tmp/rpm-tmp.54042: fg: no job control 18:01:20< Nesh> error: Bad exit status from /var/tmp/rpm-tmp.54042 (%build) 18:01:32>> taxcollector [~taxcollec@192.16.167.161] has joined #vserver 18:01:41< Bertl> try to copy the complete output, and provide an url ... 18:01:47< Bertl> hi taxcollector! 18:01:52< taxcollector> Howdy! 18:01:58< Bertl> okay Napalm, any questions left? 18:02:07< Napalm> ohhh loads 18:02:08< Napalm> :D 18:02:26< Bertl> anything unanswered? 18:02:56< Napalm> whats the recommended linux-vserver version to use? 18:03:05< Nesh> http://www.dineshmistry.com/binutil.txt 18:03:11< Nesh> wait 18:03:31< Nesh> http://www.dineshmistry.com/binutil.txt 18:04:05< Bertl> hmm, stupid rh, okay replace the %make -j 5 18:04:13< Bertl> by just make 18:04:28< Bertl> (in the spec file) 18:04:30< Nesh> here goes again 18:04:36< Nesh> compiling again 18:05:25< Nesh> http://www.dineshmistry.com/binutil.txt 18:05:27< Nesh> error 18:06:00< Bertl> didn't I say 'just' make? no %make please! 18:06:04< Nesh> oh 18:06:05< Napalm> shall i wait until you and nash have finished working your problems out 18:06:13< Nesh> forgot 18:06:15< Nesh> the % 18:06:17< Bertl> no that would take the whole night! 18:06:21< Napalm> lol 18:06:24< Nesh> LOL yes i am slow! 18:06:29< Nesh> go around me 18:06:31< Nesh> ;) 18:06:32< Bertl> just ask, I'm multitaskable ... 18:06:54< Napalm> oh dear, find yourself a extra processor nash 18:07:03< Napalm> double the power 18:07:10< Nesh> heh 18:07:12< Napalm> poor i know 18:09:14< Bertl> Napalm: there are/were some UI projects for Linux-VServer (using php and such) 18:09:57< Napalm> i saw a link on the linux-vserver site to a SF project but no files have been released i might get in contact with the admins 18:10:13< Napalm> ok 2 main questions 18:10:35< Bertl> go ahead! 18:10:40< Nesh> i got a error but I am fixing it 18:10:47< Nesh> %make DESTDIR=$RPM_BUILD_ROOT install in SPEC file 18:10:50< Nesh> makeing it make DESTDIR=$RPM_BUILD_ROOT install 18:10:53< Nesh> ok? 18:10:58< Bertl> good boy ;) 18:11:01< Nesh> yay! 18:11:21< Napalm> Bert: 1. Unification, How do i implement this? I've downloaded the RH8 image file from JVDS 18:11:52< Napalm> Bert: 2. Mounting vserver drive space so that I can limit the disk space usage? 18:11:54< Bertl> depends on the branch, but with alpha utils, you do not need to ... it's already implemented ... 18:13:07< Napalm> Bert: but the image files need to be mounted and in read-only mode 18:13:23< Bertl> hmm, why? 18:13:58< Bertl> ad 2) there is something called Context Disk Limits, which probably does what you want ... 18:14:57< Nesh> Wrote: /usr/src/redhat/SRPMS/binutils-i386-2.15.90.0.3-3mdk.src.rpm 18:14:57< Nesh> Wrote: /usr/src/redhat/RPMS/i386/binutils-i386-2.15.90.0.3-3mdk.i386.rpm 18:15:34< Bertl> good! 18:15:46< Bertl> Nesh: rpm -ql -p /usr/src/redhat/RPMS/i386/binutils-i386-2.15.90.0.3-3mdk.i386.rpm 18:15:53< Bertl> will show the files it contains ... 18:16:05< Nesh> correcto! 18:16:05< Bertl> should not interfere with any of your current files ... 18:16:15< Napalm> Bert: i will run several vservers from one image file, and when a vserver internal user tries saving a file the unification should unlink the file and save it to the actual /vserver// space 18:16:34< Bertl> well, that's not how unification works ... 18:16:40< Napalm> but the whle unification idea works from hard linking 18:16:49< Nesh> Bertl do not appear to 18:16:56< Nesh> I ls -al some of them 18:17:00< Nesh> and they do not exist 18:17:05< Bertl> Napalm: what you think is cow and transparent overlay 18:17:13< Bertl> Nesh: oaky install it 18:17:31< Napalm> Bert: if i was to mount the image then use the clone option to clone with unification making the hard links to the mounted image file would that work? 18:17:32< Nesh> 1:binutils-i386 installed 18:17:37< Bertl> http://vserver.13thfloor.at/Stuff/Cross/gcc-i386-3.4.0-4mdk.src.rpm 18:17:51< Bertl> you have to 'adapt' the spec file similar to the binutils one 18:17:59< Bertl> and compile it the same way 18:18:02< Nesh> ok 18:18:22< Bertl> Napalm: no because that isn't the way hard links work ;) 18:19:02< Napalm> hard links are entries in the FAT table that point to the same drive area where the actual data is stored yea? 18:19:20< Bertl> what you can do is: copy the image onto your harddisk, duplicate the whole tree and unify it with the original 18:19:23< Nesh> rpmbuild -ba /usr/src/redhat/SPECS/gcc-cross.spec --define "CROSS_ARCH i386" 18:19:50< Bertl> Napalm: dos is FAT 18:20:08< Napalm> so if i copy the files from the mounted image file to /vservers/rh8_base/ 18:20:08< Bertl> Nesh: yes 18:20:27< Bertl> Napalm: for example, then you can do hardlinks to that ... 18:20:30< Napalm> i should be able to choose clone and unification for that 18:20:35< Bertl> yes 18:20:47< Napalm> seems so simple now, soh 18:20:50< Napalm> *doh 18:21:15< Bertl> alpha util-vserver will do that for you, and allow to create the 'original' out of thin air ... 18:21:45< Nesh> i have a feeling this compile is going to take a while. 18:22:05< Bertl> Napalm: I'd suggest using them as backend anyways, because when you are finished, the alpha tools are probably stable ... 18:22:16< Napalm> util-vserver-0.29.4-1mdk.i586.rpm 18:22:16< Napalm> util-vserver-linuxconf-0.29.4-1mdk.i586.rpm 18:22:21< Bertl> Nesh: well, not that long 18:22:36< Nesh> ok 18:22:38< Napalm> Bert: i installed those RPMs 18:22:48< Bertl> I know ... 18:22:54< Bertl> http://www.linux-vserver.org/index.php?page=alpha+util-vserver 18:23:23< Bertl> http://www-user.tu-chemnitz.de/~ensc/util-vserver/doc/virtual-servers.pdf 18:23:39< Bertl> (german) 18:26:41< Napalm> yep i've got it but im an english only person 18:26:54< Bertl> okay, then forget the second one ;) 18:27:10< Napalm> i tried to interpret the presentation before 18:27:15< Napalm> was a little helpful 18:27:27< Nesh> Wrote: /usr/src/redhat/SRPMS/gcc-i386-3.4.0-4mdk.src.rpm 18:27:27< Nesh> Wrote: /usr/src/redhat/RPMS/i386/gcc-i386-3.4.0-4mdk.i386.rpm 18:28:03< Bertl> check, install ... 18:28:27< Nesh> yeah creates like 18:28:31< Nesh> /usr/bin/i386-linux-cpp 18:28:35< Nesh> /usr/bin/i386-linux-gcc 18:28:38< Nesh> which do not exisit 18:29:30< Nesh> rpm -Uvh --test /usr/src/redhat/RPMS/i386/gcc-i386-3.4.0-4mdk.i386.rpm 18:29:31< Nesh> Preparing... ########################################### [100%] 18:29:49< Nesh> seems ok 18:30:17< Bertl> okay ... 18:30:18< Nesh> should I -i it? 18:30:30< Bertl> I'd say yes ;) 18:30:33< Nesh> heh 18:30:42< Nesh> done. 18:30:49< Bertl> you can now use it this way: 18:31:01< Bertl> cd /path/to/kernel/source 18:31:20< Nesh> k 18:31:21< Bertl> make ARCH=i386 CROSS_COMPILE=i386-linux- 18:31:40< Napalm> Bert: which rpm does the vnameserver util come in 18:31:51< Napalm> *vnamespace 18:31:53< Bertl> Nesh: and if everything works as expected, you are cross compiling an i386 kernel 18:32:11< Napalm> Bert: because it does'nt seem to be installed 18:32:24< Bertl> yes, it's part of the alpha utils ... 18:32:47< Napalm> shall i remove those rpms and compile them manually instead 18:32:49< Nesh> so make bzImage ARCH=i386 CROSS_COMPILE=i386-linux- 18:32:54< Nesh> or? 18:33:40< Napalm> im thinking at the moment i might reinstall the whole box :s 18:34:08< Bertl> Napalm: I usually provide rpms for Mandrake ... 18:34:29< Bertl> there is one for 0.29.209 IIRC ... and you can have some for 0.29.211 ... 18:34:34< Nesh> http://www.dineshmistry.com/cross-compile-error.txt 18:35:23< Napalm> im using RH9 for the main server 18:35:28< Napalm> :s i know i know 18:35:38< Napalm> its just for testing purpurses 18:36:02< Napalm> Bert: Downloading util-vserver/alpha/util-vserver-0.29.211.tar.bz2 18:37:20< Nesh> Hmm 18:37:53< Bertl> btw, what is make modules? 18:38:09< Nesh> makes the modules for the kernel? 18:38:10< Napalm> Bert: is it possible to run FreeBSD vserver units? 18:38:16< Bertl> Nesh aren't you compiling a 2.6 kernel? 18:38:24< Nesh> yeah 18:38:30< Nesh> is that wrong for 2.6 18:38:45< Bertl> so modules will be included by 'make' 18:38:49< Nesh> oh 18:39:06< Nesh> no need for modules modules_install bzImage anymore? 18:39:17< Bertl> Napalm: well, if you manage to get userspace bsd utils which run with linux kernel, sure ... no not really 18:39:21< taxcollector> Napalm: To build an rpm using util-vserver-0.29.211.tar.bz2, install the dietlibc-0.24-4 rpm from Fedora, then do a 'rpmbuild -ta util-vserver-0.29.211.tar.bz2', then do an 'rpm -i' of the resulting rpms. 18:39:52< Bertl> Nesh modules_install yes, the rest, no 18:40:14< Nesh> http://www.dineshmistry.com/cross-compile-error.txt 18:40:42< Bertl> hmm, maybe your make is too old too ... 18:40:50< Napalm> Taxcollector: im compiling the utils now? would you suggest i make rpms? 18:41:21< Nesh> GNU Make version 3.79.1, by Richard Stallman and Roland McGrath. 18:41:44< Bertl> # make --version 18:41:44< Bertl> GNU Make 3.80 18:41:52< Bertl> but 3.79 should work ... 18:42:44< Napalm> Bert: i know this may seem like a dumb sort of question, what OS can be used with the vservers i was under the impression that its possible to install any linux flavored distro? 18:43:39< Nesh> lemme upgrade from make-3.79.1-8 to make-3.79.1-18 18:44:25< taxcollector> Napalm: I prefer managing the files with rpms; plus when you build and install using the rpms the usual util-vserver paths are automatically used. 18:44:31< Nesh> uhg nope 18:44:47< taxcollector> At least the ones people using the mandrake rpms will be familiar with 18:45:01< Nesh> build 3.8 by hand 18:45:59< Nesh> bert same error 18:46:15< Nesh> http://www.dineshmistry.com/cross-compile-error.txt 18:46:17< Napalm> Bert: i just thought how about this. Install FreeBSD on the main server with the kenel patches then install the linux emulation kernel module i should be able to run the tools and have vservers in freebsd? or am i barking up the wrong tree 18:46:52< taxcollector> Napalm: Wrong tree. 18:47:06< Napalm> Taxcollector: im not yet familiar with the Fedora takeover of the Linux support, can i just install Fedora rpms on RH9? 18:47:16< Bertl> Napalm: well, you have to adjust the patches a little .... 8-) 18:47:37< taxcollector> Napalm: In this case, it works. Generally, you can't make that assumption though. 18:47:41< Napalm> Bert: would it be possible for you to help adjust the patch? 18:47:45< Napalm> :D 18:48:12< Bertl> Napalm: sure, wat a moment I'm just fetching my bank account number ... 18:48:24< Napalm> ok how much are we talking? 18:48:37< Bertl> money or work? 18:48:48< Napalm> both, pm me if you want it to be kept private 18:49:11< Bertl> np, well, it's basically an implementation from scratch 18:49:20< Napalm> oh 18:49:22< Napalm> :s 18:49:32< Bertl> and re-implementing linux-vserver took me about 3 months ... 18:50:01< Napalm> cost involved? 18:50:08< Bertl> so I'd says 2.5 months without testing, and about 3 month to get used to the bsd kernel ... 18:50:55< Nesh> i hate rh7.3 18:51:02< rs> re 18:51:06< Bertl> Napalm: I usually charge about 80-120 EUR per hour ... you can do the math ... 18:51:06< Nesh> re 18:51:11< Bertl> hi rs! 18:51:22< Napalm> hmm, too much at the moment 18:51:24< Napalm> ;) 18:51:35< Bertl> so you have to stay with linux ... 18:51:48< Bertl> btw, what would be the advantage in your opinion anyway? 18:52:13< rs> was a meeting day today :/ 18:52:23< Napalm> the security aspects of the FreeBSD OS 18:52:35< rs> sorry I didn't got time to test the pre14 18:52:38< Bertl> you mean NetBSD, not FreeBSD 18:52:46< Napalm> yes 18:52:56< rs> I have a node with it if you want, I can put your vserver on, do you want ? 18:53:00< rs> it 18:53:39< Bertl> rs: well, no need to hurry ... I just want to know if we can go devel this week ... so I'm happy about feedback ... 18:53:58< rs> I hope we can 18:54:37< Bertl> btw, could you re-test the kmsg stuff, it works here with pre14 (no modifications on that though) 18:55:09< rs> ok, wait a minute 18:57:18< Bertl> # ls /proc/kmsg 18:57:18< Bertl> /proc/kmsg 18:57:18< Bertl> # chcontext --ctx 100 ls /proc/kmsg 18:57:18< Bertl> New security context is 100 18:57:18< Bertl> /proc/kmsg 18:57:21< Bertl> --- 18:57:27< Bertl> # setattr --hide /proc/kmsg 18:57:27< Bertl> # chcontext --ctx 100 ls /proc/kmsg 18:57:27< Bertl> New security context is 100 18:57:27< Bertl> ls: /proc/kmsg: No such file or directory 18:58:21< rs> debian:~# cat /proc/kmsg 18:58:21< rs> cat: /proc/kmsg: Operation not permitted 18:58:21< rs> debian:~# ls /proc/kmsg 18:58:21< rs> /proc/kmsg 18:58:21< rs> debian:~# ls /proc/|grep kmsg 18:58:24< rs> debian:~# 18:58:40< rs> same behavious 18:58:53< Bertl> strange ... what is the difference? 18:59:10< rs> hmm I use namespace ? 18:59:20< Bertl> could you try with my commands? 18:59:25< rs> sure 18:59:46< rs> bvdsn01:~# setattr --hide /proc/kmsg 18:59:46< rs> bvdsn01:~# chcontext --ctx 100 ls /proc/kmsg 18:59:46< rs> New security context is 100 18:59:46< rs> /proc/kmsg 19:00:07< rs> what is you util-vserver version btw ? 19:00:22< Bertl> 0.29.209 ... 19:00:36< rs> same version for me 19:00:41< rs> really strange 19:01:13< Bertl> okay, give pre14 a try, and see if it's still there ... 19:01:26< rs> it is the pre14 19:01:34< Bertl> uhm ... 19:01:49< Bertl> ahh! 19:02:06< Bertl> your klogd is probably reading that ... 19:02:20< Bertl> could you shut down the host klogd for a test? 19:02:25< rs> yep 19:02:42< Nesh> bert any ides? 19:02:44< rs> bvdsn01:~# /etc/init.d/klogd stop 19:02:44< rs> Stopping kernel log daemon: klogd. 19:02:44< rs> bvdsn01:~# chcontext --ctx 100 ls /proc/kmsg 19:02:44< rs> New security context is 100 19:02:44< rs> ls: /proc/kmsg: No such file or directory 19:02:50< rs> you got it :) 19:02:52< Bertl> hehe! 19:03:21< Bertl> Nesh: get another distro, no sorry ... would need deeper investigation, we can try later ... 19:03:29< Nesh> okok 19:03:33< Nesh> thanks1 19:03:35< Nesh> er! 19:03:46< Bertl> rs: the question is, can we work around that ... 19:04:10< rs> good question 19:04:37< rs> running without klogd on the host is a little annoying :) 19:04:37< Bertl> maybe returning a ESRCH instead of an EPERM would be sufficient ... 19:05:31< rs> maybe 19:06:05< rs> but it will not be reachable even if we didn't hiden it isn't it ? 19:06:32< Bertl> no, and I'm not sure that hiding it would help the guest in any way ;) 19:06:49< Bertl> so it would be interesting to test this first ;) 19:06:57< rs> yes 19:07:21< rs> I can do that tomorrow if you want, I have to go 19:07:27< Bertl> np, cya 19:07:41< rs> see you tomorrow :) 19:07:48>> rs [rs@ice.aspic.com] has quit [Quit: bye] 19:08:20< Nesh> Bertl I should be able to build on another machine and install on rh7.3 machine? 19:08:32< Bertl> the kernel? yes sure! 19:08:36< Nesh> ah 19:09:54< taxcollector> Nesh: Why are you still on 7.3? 19:10:02< Nesh> too hard to upgrade 19:10:07< Nesh> too much stuff going on I mean 19:10:15< taxcollector> I ask because when you get to the point of compiling util-vserver there will be more grief 19:10:20< Nesh> need to build a parallel machine and migrate 19:10:32< Nesh> its production. 19:10:38< Bertl> taxcollector: grief? you must be joking? 19:11:04< taxcollector> I tried it once -- there was a lot of dependencies that wasn't resolved with a simple download and rpm -i 19:11:09< Bertl> this will be deep deep sadness .... 19:11:26< Nesh> is there a quick and dirty way to build a .RPM of the kernel from /usr/src/linux ? 19:11:28< taxcollector> I don't have a problem compiling and scrounging around for code 19:11:46< taxcollector> but my impression is that for Nesh another path would be easier. 19:12:26< Bertl> I agree, it would be best to upgrade to Mandrake-10.0 and be happy ... 19:12:37< Nesh> Mandrake? 19:12:39< Nesh> hmm 19:13:00< Bertl> well, it's rpm based, so no new tools to learn (at least not that many) 19:13:24< Nesh> ahh 19:13:29< Nesh> at home I run gentoo 19:13:48< Nesh> so any anwser to my previos question? 19:17:57< Nesh> make rpm 19:17:57< Nesh> ! 19:17:57< Bertl> building RPMs of kernels (especially patched ones) isn't such a good idea ... 19:17:57< Nesh> but i want to be able to move it to another server? 19:17:57< Nesh> just cp the bzImage and modules etc? 19:17:57< Bertl> yep ... 19:17:57< Nesh> hmm 19:17:57< Bertl> don't forget, 2.6 kernel requires module-inst-utils ... (not modutils) 19:17:57< Nesh> oh yeah I installed that 19:17:58< Nesh> but I dun build a modular kernel 19:18:02< Nesh> monolithic 19:18:21< Bertl> so you have no modules then ... 19:18:26< Nesh> correct 19:18:34< Bertl> so why do you want to copy them over? 19:18:41< Nesh> [root@host01:/home/admin/Docs]$ /sbin/lsmod 19:18:41< Nesh> Module Size Used by Not tainted 19:18:41< Nesh> [root@host01:/home/admin/Docs]$ 19:18:50< Nesh> i guess I do not need to 19:18:51< Nesh> :) 19:24:02< Nesh> is the 2.6 kernel faster benchmark wise then 2.4 i wonder 19:24:57< Bertl> probably depends on the benchmark ... 19:25:12< Bertl> but the kernel per se improved a lot ... 19:26:11< Nesh> i saw a few benchmarks where it did not at somethings and slowed down at others. 19:28:11< Bertl> okay, leaving now .. cu later ... 19:28:22< Nesh> byebye 19:28:27>> Bertl is now known as Bertl_oO 19:43:10>> shuri [~shushushu@cpu183.adsl.qc.bellglobal.com] has joined #vserver 19:46:17< Napalm> can anyone help with a new server setup 19:46:36< Napalm> installed kernel 2.4.26-vs1.27 19:47:06< taxcollector> Which tools do you have installed now? 19:47:39< Napalm> util-vserver-0.29.211-0 from your RPMs 19:48:30< taxcollector> OK -- for the stable patch I would go back to util-vserver-0.29.4 19:48:45< taxcollector> Uninstall those rpms with 'rpm -e' 19:49:12< taxcollector> And then try http://www.13thfloor.at/vserver/s_release/v1.27/util-vserver-0.29.4-1mdk.i586.rpm 19:49:58< Napalm> http://www.vhd.org.uk/vserver-stop-log.txt 19:50:05< Napalm> thats the stop log with the errors 19:50:11< Napalm> every starts fine thou 19:51:03< taxcollector> That one is in the FAQ 19:51:04< taxcollector> http://www.linux-vserver.org/index.php?page=Linux-Vserver+FAQ 19:51:18< taxcollector> Search text for hwclock 19:56:23< Napalm> ok sorted that one 19:56:34< Napalm> now we have http://www.vhd.org.uk/vserver-stop-log2.txt 19:59:14< taxcollector> Don't have any idea about that 20:04:10< Napalm> ok i think ive sorted most of these problems out 20:04:17< taxcollector> Cool 20:10:20< taxcollector> Leaving now, catch ya later 20:10:43>> taxcollector is now known as taxcollector_oO 20:12:48>> taxcollector_oO [~taxcollec@192.16.167.161] has quit [Quit: ] 20:21:07< Nesh> dav 20:26:11>> nalfein [~gaertner@212.68.83.129] has quit [Quit: Serverwechsel] 20:31:32>> monrad [~monrad@213083190234.sonofon.dk] has joined #vserver 21:04:14>> shuri [~shushushu@cpu183.adsl.qc.bellglobal.com] has quit [Quit: http://base2091.com] 21:09:22< Napalm> does anyone know if i can use util-vserver alpha with vserver kernel patch 1.27 21:09:23< Napalm> ? 21:10:07< Doener`> IIRC that should be possible 21:10:29< Napalm> IIRC? 21:10:38< Napalm> ok well i'll have ago 21:10:54< Napalm> are the rpms for the alpha version? 21:25:04< Doener`> IIRC = if i remember correctly (IIRC ;)) 21:25:52< Doener`> AFAIK there are no rpms, but there's an uninstall target in the makefile 21:31:41< Napalm> cool i'll try and compile then 21:36:33>> Illegal [~noone@200-103-242-126.ctame7042.dsl.brasiltelecom.net.br] has joined #vserver 21:39:59>> lexo_ [~LeXo@lns-th2-4f-81-56-252-185.adsl.proxad.net] has joined #vserver 21:46:57>> franck [~LeXo@lns-th2-4f-81-56-252-185.adsl.proxad.net] has quit [Ping timeout: 480 seconds] 22:05:58>> Apollo [~throwaway@caracal.norcomcable.ca] has quit [Quit: ] 22:24:36 * Illegal : estourando os timpanos com - Drowning Pool - Cast Me Aside - [Musica 83] [sCrIpInTaO 20.0 by UZY] 22:30:33>> Illegal [~noone@200-103-242-126.ctame7042.dsl.brasiltelecom.net.br] has quit [Quit: [BX] You can /quit any time you like, but you can never leave] 22:47:22>> axu [gl@83-64-20-69.dynamic.home.xdsl-line.inode.at] has joined #vserver 22:47:26< axu> hi folks 22:48:11>> axu [gl@83-64-20-69.dynamic.home.xdsl-line.inode.at] has quit [Quit: ] 22:49:45>> axu [gl@83-64-20-69.dynamic.home.xdsl-line.inode.at] has joined #vserver 22:49:48< axu> ups re :) 22:50:14< axu> the problem with clicking is the clicking :) 23:06:35>> monrad [~monrad@213083190234.sonofon.dk] has quit [Quit: Leaving] 23:26:41>> Bertl_oO is now known as Bertl 23:26:47< Bertl> hi everyone! 23:28:37< kestrel> hey herbert 23:28:57< Bertl> hi alec! how's life? 23:38:40< kestrel> not bad 23:38:43< kestrel> how are you doing? 23:39:05< Bertl> fine, thanks ... so far so good ;) 23:39:39< kestrel> what's news with vserver? any cool developments? 23:39:54< Bertl> well, what's you last status? 23:40:36< Bertl> maybe my paper is new for you? 23:40:38< kestrel> vs1.27! :) 23:40:50< kestrel> yep, where is it? 23:40:56< Bertl> http://vserver.13thfloor.at/Stuff/PAPER-05.1.txt 23:41:18< Bertl> it's not the final version, but not far from it ... 23:41:37< kestrel> aha, cool! 23:41:55< kestrel> loooong, it will take me a while to read it (i'm at work) 23:42:07< kestrel> when did you write that? 23:42:11< Bertl> well, no need to read it now ;) 23:42:15< kestrel> is it for a conference? 23:42:31< Bertl> it's the LinuxTag paper ... 23:42:43< kestrel> aha 23:42:52< Bertl> (some gathering in germany ;) 23:43:19< kestrel> yeah, i have heard of linuxtag before i think 23:43:25< kestrel> hey, check this out herbert: http://projects.edgewall.com/trac 23:43:37< kestrel> i found it yesterday, looks interesting 23:44:02< Doener`> hmm... can anybody tell me what events/0 is good for? 23:44:13< Bertl> in /dev ? 23:44:19< Doener`> in ps output ;) 23:44:36< Bertl> ps like PostScript? 23:44:46< cgone> Doener`: it's a kernel thread 23:45:01< cgone> Doener`: it's used for scheduling work in a process context 23:45:01< Doener`> ps like ps auxww ;) 23:45:29< Bertl> damn, I was hoping to confuse you! ;) 23:46:16< Bertl> btw, shouldn't kernel threads have some brackets around? 23:46:25< Napalm> hey Bert 23:46:52< Doener`> Bertl: sorry, forget to paste them ;) 23:46:57< Bertl> hi Napalm, hope you are still away from the spicey food ;) 23:47:24< Napalm> spiecey food is toooo hot for me ;) 23:47:39< Doener`> hmm... no it's gone... there was a rather busy mysql query and there were about 50 [events/0] thingies in ps output and a nice load average of about 100 23:47:44< Doener`> s/no/now/ 23:48:08< Bertl> 2.6? 23:48:14< Doener`> yepp 23:48:28< Doener`> not vserver-related btw 23:48:43< Bertl> there are a lot of 'new' kernel threads and helpers ... 23:49:05< Napalm> Bert: if you use the upyime command in a vserver is it actuall giving the stats for the vserver or the main server? 23:49:29< Bertl> depends on your version and flags ... 23:49:52< Napalm> would'nt it be good to make a vuptime command, will give you the uptime for all the vservers in a ascii table 23:49:54< Bertl> devel + exp has uptime virtualization ... 23:50:20< Bertl> Napalm: hmm, what about using vserver-stat for now? 23:51:02>> serving [~serving@213.186.189.95] has quit [Read error: Connection reset by peer] 23:51:10< Napalm> Bert: how about 2.4.26-vs1.27 and util-vserver-0.29.211? 23:51:28< Napalm> Bert: yer vserver is fine but no load averages 23:51:33< Bertl> well, should work ... but it's basically untested ... 23:51:49< Bertl> Napalm: there is no loadavg virtualization yet 23:52:19< Napalm> Bert: im suggesting it for future developement 23:52:48< Bertl> loadavg virt is already on my todo list ... I'm just waiting for it to level up ;) 23:53:26< Napalm> cool 23:53:29< Napalm> ;) 23:53:33< Bertl> but keep them idears coming, I'm sure there will be something new ... 23:53:55< Bertl> (or something I already forgot ;) 23:54:27< Bertl> btw, I'm always happy about people _testing_ those features ... 23:55:23< Bertl> so if you start testing stuff, and are especially interested in, let's say loadavg virtualization ;), then I'm inclined to give it higher priority 23:55:55< Napalm> ive just compiled and removed the old utils but now the new installed ones seem to look for /usr/local/etc/vservers/ instead of /vservers/ as the VSERVERROOT 23:56:01< axu> hi bertl :) 23:56:07< Bertl> evening axu! 23:56:40< Bertl> axu: are you doing nightshifts, or did you become a true vserver fan? 23:56:46< Bertl> (or both ;) 23:56:48< axu> hehe 23:56:56 * axu s @ home 23:56:59< Napalm> once i have a stable setup together i will use it as a "todo" when i setup the dedicated box, then i can use my testing box to test future releases 23:57:15< Doener`> Napalm: what does 'which vserver' say? 23:57:34< axu> Bertl, nightshifts what i call the stuff that has to do with work :) this is mor like small party ;) 23:57:42< Bertl> Napalm: get an overview with vserver-info - SYSINFO 23:58:01< Napalm> Donener: I feel so stupid now, its "/usr/local/sbin/vserver" 23:58:03< Bertl> axu: ahh small party, 2 people ... vserver .. I see ... 23:58:24< Doener`> Napalm: ok, there should be a file util-vserver-vars in the same folder, right? 23:58:43< axu> Bertl, you see what you want to :) 23:59:02< Doener`> Napalm: IIRC you can configure the prefix while building, but we should be able to change it without a re-build 23:59:17< Bertl> btw, did enrico show up? --- Log closed śro maj 05 00:00:00 2004