[00:18] Nick change: Bertl_oO -> Bertl [00:25] JonB (~NoSuchUse@129.142.112.33.ip.tele2adsl.dk) joined #vserver. [00:31] Bertl (~herbert@MAIL.13thfloor.at) got netsplit. [00:31] maharaja (maja@ipax.tk) got netsplit. [00:31] deadguy (deadguy@bananajoe.big.du.se) got netsplit. [00:31] pollar (bc3@dievai.net) got netsplit. [00:31] _Axu_ (~axu@mail.mcmarketing.at) got netsplit. [00:31] Doener_zZz (~doener@p5082DA83.dip.t-dialin.net) got netsplit. [00:31] riel (~riel@riel.netop.oftc.net) got netsplit. [00:31] talon (talon@host-63-149-223-100.irwinresearch.com) got netsplit. [00:31] mcp (~hightower@wolk-project.de) got netsplit. [00:31] Zoiah (Zoiah@matryoshka.zoiah.net) got netsplit. [00:33] maharaja (maja@ipax.tk) returned to #vserver. [00:33] deadguy (deadguy@bananajoe.big.du.se) returned to #vserver. [00:33] pollar (bc3@dievai.net) returned to #vserver. [00:33] _Axu_ (~axu@mail.mcmarketing.at) returned to #vserver. [00:33] mcp (~hightower@wolk-project.de) returned to #vserver. [00:33] Doener_zZz (~doener@p5082DA83.dip.t-dialin.net) returned to #vserver. [00:33] riel (~riel@riel.netop.oftc.net) joined #vserver. [00:33] talon (talon@host-63-149-223-100.irwinresearch.com) returned to #vserver. [00:33] Zoiah (Zoiah@matryoshka.zoiah.net) returned to #vserver. [00:34] Bertl (~herbert@MAIL.13thfloor.at) returned to #vserver. [00:56] Doener_aw (~doener@p5082DBF5.dip.t-dialin.net) joined #vserver. [00:56] hi Doener! [00:56] hi! [00:56] Nick change: Doener_aw -> Doener [00:56] i wonder when my isp will have solved this... [00:57] the minute you change your isp ;) [00:57] solved what ? [00:57] i guess in the last two weeks i've been unable to connect for about one day in total... [00:57] whatever murphy doesn't care that much ;) [00:58] don't know what exactly is the problem... whenever i call them they have a different issue... [01:00] Doener_zZz (~doener@p5082DA83.dip.t-dialin.net) left irc: Ping timeout: 480 seconds [01:00] I had this once, the company fired their best man, and nobody knew how the stuff actually worked before ... [01:00] explanations ranged from defective UPS to bad firmware in switch ... [01:01] Bertl: sorry you got fired [01:01] hehe [01:02] ;) [01:03] JonB, so let's talk about one of your many projects %-) [01:04] Bertl: well, it seems like i almost have 2 projects closed [01:05] Bertl: one is with a friend, the other alone [01:05] the alone one is making linux run on a embedded vr4133 board [01:06] which hasn't been done before, right? [01:06] and making a bootloader that can choose between 2 kernel + file systems, so it can be updated [01:07] like grub? [01:07] and possibly adding support in the kernel for using the cryptografic stuff in the cpu [01:07] Bertl: possibly, but i was thinking of assembler, since they want to save flash space [01:08] Bertl: the other is a blind reverse engineering project with a friend [01:08] i read the freebsd GEOM-GDBE driver, and write a documentation specs. [01:08] then my friend read what i created, and create a linux driver [01:09] hmm sounds interesting ... [01:10] we want to try how hard reverse engineering is [01:10] I would suggest 'Paycheck' for educational purposes ;) [01:11] paycheck =? [01:11] movie about reverse engineering ... [01:13] but it is with ben affleck :( [01:14] and?, and? ... [01:15] i've seen better acting at goatse.cx [01:16] the only good movies he's in, are movies where matt daemon is as well [01:17] you missed the Uma Thurman part, but whatever you say ... [01:17] Bertl: too old :/ [01:18] hrmpf, she is my age ... am I too old too? [01:18] i'm not into guys [01:18] hehe okay, you win! [01:18] wooohoo [01:18] what do i get ? [01:19] a shrubbery! [01:19] actually somebody owes me two shrubberies IIRC, so he now owes you one ... [01:19] cool :) [01:20] but you have to collect it yourself ... [01:20] hmm :( [01:24] Bertl: how old are you anyway ? [01:25] well, it's very close to Uma ... [01:25] then you are my age [01:27] and thats not old ;-P [01:27] but i still find uma too old [01:29] Nick change: riel -> unriel [01:46] JonB (~NoSuchUse@129.142.112.33.ip.tele2adsl.dk) left irc: Quit: Leaving [02:27] paul (~irssi@195.202.59.212) left irc: Quit: leaving [02:54] butonic (~butonic@port-212-202-25-66.reverse.qsc.de) joined #vserver. [02:54] hi butonic! [02:55] hi everybody [02:57] I want to set up three firewalls each in its own vserver. However, I dont know how to set the filter rules for each of them ... well I do not even know whether this is possible with vlinux. Any suggestions? [02:57] err vserver that is [02:58] well, you have to put all rules on the host, but besides this, it should not be a problem ... [02:58] it will be a good idea to ahve separate routing tables for them ... [02:59] Doener_zZz (~doener@pD9E12F9E.dip.t-dialin.net) joined #vserver. [03:00] so it should suffice to set up three subnets? I had prefered to seperate the firewall setup for each vserver, but i assume everything has to be configured on the host, hasn't it? [03:01] as I said, yes, unless you use FreeVPS which has per vserver iptables (but I do not know how secure this is) [03:01] Nick change: Doener_zZz -> Doener_ [03:01] Doener (~doener@p5082DBF5.dip.t-dialin.net) left irc: Ping timeout: 480 seconds [03:02] hmm, ok thx a lot! [03:02] np [03:02] ill report to the wiki when i got my DMZ on a vserver-host ;) - n8 [03:02] butonic (~butonic@port-212-202-25-66.reverse.qsc.de) left #vserver. [03:07] Zoiah (Zoiah@matryoshka.zoiah.net) got netsplit. [03:07] talon (talon@host-63-149-223-100.irwinresearch.com) got netsplit. [03:07] _Axu_ (~axu@mail.mcmarketing.at) got netsplit. [03:07] pollar (bc3@dievai.net) got netsplit. [03:07] deadguy (deadguy@bananajoe.big.du.se) got netsplit. [03:07] maharaja (maja@ipax.tk) got netsplit. [03:07] unriel (~riel@riel.netop.oftc.net) got netsplit. [03:07] mcp (~hightower@wolk-project.de) got netsplit. [03:07] Bertl (~herbert@MAIL.13thfloor.at) got netsplit. [03:07] youam (~youam@ciara.youam.de) got netsplit. [03:07] Doener_ (~doener@pD9E12F9E.dip.t-dialin.net) got netsplit. [03:07] kestrel (athomas@home.swapoff.org) got netsplit. [03:07] eyck (~eyck@62.233.189.138) got netsplit. [03:07] ccooke (~ccooke@spc1-walt1-4-0-cust238.lond.broadband.ntl.com) got netsplit. [03:07] sladen (paul@starsky.19inch.net) got netsplit. [03:07] Medivh (ck@62.93.217.199) got netsplit. [03:07] kestrelw (~athomas@o2rosock0a.optus.net.au) got netsplit. [03:07] cdub (~chrisw@fw.osdl.org) got netsplit. [03:07] lilo (levin@lilo.usercloak.oftc.net) got netsplit. [03:07] Doener_ (~doener@pD9E12F9E.dip.t-dialin.net) returned to #vserver. [03:07] kestrel (athomas@home.swapoff.org) returned to #vserver. [03:07] eyck (~eyck@62.233.189.138) returned to #vserver. [03:07] ccooke (~ccooke@spc1-walt1-4-0-cust238.lond.broadband.ntl.com) returned to #vserver. [03:07] sladen (paul@starsky.19inch.net) returned to #vserver. [03:07] lilo (levin@lilo.usercloak.oftc.net) returned to #vserver. [03:07] kestrelw (~athomas@o2rosock0a.optus.net.au) returned to #vserver. [03:07] cdub (~chrisw@fw.osdl.org) returned to #vserver. [03:07] Medivh (ck@62.93.217.199) returned to #vserver. [03:08] Bertl (~herbert@MAIL.13thfloor.at) returned to #vserver. [03:08] maharaja (maja@ipax.tk) returned to #vserver. [03:08] deadguy (deadguy@bananajoe.big.du.se) returned to #vserver. [03:08] pollar (bc3@dievai.net) returned to #vserver. [03:08] _Axu_ (~axu@mail.mcmarketing.at) returned to #vserver. [03:08] mcp (~hightower@wolk-project.de) returned to #vserver. [03:08] unriel (~riel@riel.netop.oftc.net) returned to #vserver. [03:08] talon (talon@host-63-149-223-100.irwinresearch.com) returned to #vserver. [03:08] Zoiah (Zoiah@matryoshka.zoiah.net) returned to #vserver. [03:08] youam (~youam@ciara.youam.de) returned to #vserver. [03:51] Mister_A_ (~mab@nat01-clo-ext.Rutgers.EDU) joined #vserver. [04:09] Mister_A_ (~mab@nat01-clo-ext.Rutgers.EDU) left irc: [04:38] polarfox (bc3@dievai.net) joined #vserver. [04:39] pollar (bc3@dievai.net) left irc: Read error: Connection reset by peer [05:00] impact (impact@z10n.tech.us.edu.pl) joined #vserver. [05:00] hi impact! [05:01] hi :) [05:16] soor (as@pD951AABC.dip.t-dialin.net) joined #vserver. [05:16] hi soor! [05:18] soor_ (as@217.81.171.67) left irc: Ping timeout: 499 seconds [05:23] impact (impact@z10n.tech.us.edu.pl) left irc: Quit: Lost terminal [05:26] Nick change: cdub -> cgone [05:53] ben_ (ben@bengrimm-host229.dsl.visi.com) joined #vserver. [05:54] kestrel (athomas@home.swapoff.org) left irc: Quit: brb [05:56] kestrel (athomas@home.swapoff.org) joined #vserver. [06:02] Bertl: have any 'major' changes been made to the way a vserver is actually started/stopped, i.e. the final call of the vserver bash script? [06:02] say from vs1.00 to 0.09.8... [06:03] hum, what is the 'final' call? [06:04] that quite long line... 'chroot...chbind...chbind...whatever' [06:04] i guess thats the final action that script performs... [06:05] Doener_: about which util-vserver branch are you speaking? [06:05] for alpha: yes [06:05] not specific to util-vserver but vserver in general... [06:06] i think about writing a daemon to support starting/stopping vservers using configurations stored in a database... [06:07] so i need to known if the way a vserver is started/stopped differs from one vserver version to another [06:07] Doener_: what do you mean with 'vserver version'? [06:07] the kernel patch [06:08] the concept of a 'vserver' is defined by the userspace utilities... [06:09] the kernel does not know anything about a 'vserver' [06:09] it just sees processes and contexts [06:11] okay, you're right... i even told this once to someone myself... *bangs head against wall* [06:11] then in what way it has changed in the alpha-branch? [06:11] very much ;) [06:11] and it will change more [06:12] hmm... then i may end up with less work by just converting the database stuff to a configuration suitable for util-vserver... [06:13] start/stop commands are either defined by an init-style (sysv/minit/plain currently), or by manual commands [06:13] there are different kinds for start, stop and synchronizing [06:16] guess i'll have a look at it and decide afterwards what way is more suitable... [06:21] Doener_zZz (~doener@pD9E12E6C.dip.t-dialin.net) joined #vserver. [06:21] Nick change: Doener_zZz -> Doener [06:25] Doener_ (~doener@pD9E12F9E.dip.t-dialin.net) left irc: Ping timeout: 480 seconds [06:53] night everyone! [06:54] Nick change: Bertl -> Bertl_zZ [07:41] kestrel (athomas@home.swapoff.org) left irc: Quit: reboot [07:44] Doener (~doener@pD9E12E6C.dip.t-dialin.net) left irc: Ping timeout: 499 seconds [08:34] eyck (~eyck@62.233.189.138) left irc: Ping timeout: 480 seconds [08:40] eyck (~eyck@62.233.189.138) joined #vserver. [08:47] kestrel (athomas@home.swapoff.org) joined #vserver. [08:47] hello there [09:15] hello [09:33] Doener (~doener@pD9588136.dip.t-dialin.net) joined #vserver. [09:33] youam (~youam@ciara.youam.de) got netsplit. [09:36] youam (~youam@ciara.youam.de) returned to #vserver. [10:24] ben_ (ben@bengrimm-host229.dsl.visi.com) left irc: Ping timeout: 499 seconds [11:07] kestrel (athomas@home.swapoff.org) left irc: Quit: ircII EPIC4-1.0.1 -- Are we there yet? [11:21] kestrel (athomas@home.swapoff.org) joined #vserver. [11:56] rs (rs@ice.aspic.com) joined #vserver. [11:56] hey [12:26] chaosle (~yvan@bragi.fh-brandenburg.de) joined #vserver. [12:26] hi all [12:29] etienne (~eroulland@81.80.240.90) joined #vserver. [12:29] hi [12:35] ho [12:45] i'm testing vserver [12:47] i lonnkig for script to install a more recent mdk [12:47] s/lonnkig/looking/ [12:47] as fas as i know there 's only mdk-8.2 install script [12:49] <_Axu_> hi folks [14:14] stubbsd (~stubbsd@217.206.216.194) joined #vserver. [14:14] morning all, [14:59] serving (~serving@213.186.190.121) joined #vserver. [15:25] AHTOH (~Anton@212.1.230.115) joined #vserver. [15:25] hi ppl need help [15:36] man, i need dsl [15:37] i have a question about kernel routing in vserv [15:37] do we have routes for each vserver? [15:38] no [15:38] how it works? [15:38] if i have a route to some net through the if i dont see in my context [15:38] how will it work? [15:38] i believe routes are visible to all vservers [15:39] yeah [15:39] it just works [15:39] but what with packet sending to route which is routed to ETH i dont have [15:39] eg. my vservers are all on dummy0, but default routing out of the server is via eth0 [15:39] and it works [15:40] your case i think all your ifaces are eth0:? [15:40] no, dummy0 [15:40] dummy0:foo [15:40] dummy0:bar [16:00] JonB (~NoSuchUse@129.142.112.33.ip.tele2adsl.dk) joined #vserver. [16:15] AHTOH (~Anton@212.1.230.115) left #vserver (Client exiting). [16:32] ben (~ben@bengrimm-host225.dsl.visi.com) joined #vserver. [16:42] <_Axu_> has anyone a tip on how i can test the ctx kernelpatch works ? [16:43] <_Axu_> is still got unable to switch in context security #1 and dont know how to get any further [16:44] did you install the vserver-utils package? [16:44] <_Axu_> JonB: well, i think so :) [16:45] chcontext [16:46] <_Axu_> JonB: hmm, oki [16:47] does it exist ? [16:47] <_Axu_> JonB: jepp, but i dont get the syntax do do something other then printf ing the heltext :) [16:48] what system are you on ? [16:49] <_Axu_> hmm, sparc linux [16:49] <_Axu_> debian [16:49] did you make a filesystem image ? [16:50] <_Axu_> isnt there something like chkifctxisworking.sh ? [16:50] <_Axu_> JonB: what ? no. i have a fresh woody installation, a freshcompiled kernel that is ctx patched and want to check if ctx works. i built one vserver from the root machine [16:51] can that vserver start ? [16:51] <_Axu_> no [16:51] <_Axu_> i cant do anything vserverlike :) [16:51] what happens when you type [16:51] vserver start [16:51] <_Axu_> thats why i am blabbering the same error message into tghis channel for 2 days :) [16:51] or enter ? [16:52] i havent seen that error message [16:52] <_Axu_> JonB: i cant even do a vserver-stat [16:52] <_Axu_> vserver-stat: unable to switch in context security #1 [16:52] what about vps ? [16:52] <_Axu_> so there seems not to be a rootserver.... [16:52] <_Axu_> as far as i can tell [16:53] root server is context 0 [16:53] did you talk with bertl over this ? [16:53] <_Axu_> ok# [16:53] <_Axu_> JonB: hmm, dont think so [16:54] what is your time now ? [16:55] did you email the mailing list ? [16:56] <_Axu_> 14:58 [16:56] kay, come back in the evening, bertl is usualy on by then, and he's the guru [16:56] did you patch the kernel yourself ? [16:56] <_Axu_> JonB: thanks [16:57] <_Axu_> JonB: hehe jepp, but bertl supportet me with a fast patch to the patch. so i didnt patch all bymyself ;) [16:58] okay [16:58] so you did talk to him [16:59] <_Axu_> JonB: yes [17:00] <_Axu_> JonB: sorry, bad memory, no ecc [17:00] haha [17:05] <_Axu_> ensc: is there any plans to not be that redhatdistrospecific [17:05] <_Axu_> ? [17:12] stubbsd (~stubbsd@217.206.216.194) left irc: Ping timeout: 499 seconds [17:14] ensc isnt here, is he ? [17:15] <_Axu_> JonB: doesnt look so [17:19] i don't find a page wich explain which "features" you can't use [17:19] as tcpdump [17:19] tcpdump does'nt work in a virtual box ? [17:19] etienne: depends [17:19] Nick change: unriel -> riel [17:20] depends ? [17:20] etienne: vserver uses the capability system [17:20] so ? [17:20] etienne: meaning you can give a capability to a vserver, and/or to others. [17:20] etienne: to use tcpdump you would need CAP_NET_RAW [17:20] there are others [17:20] k [17:22] thanks [17:25] riel (~riel@riel.netop.oftc.net) left #vserver (Excess food). [17:26] hmmm [17:26] another one [17:26] if a give S_CAPS="CAP_NET_RAW" to a vserver [17:27] it can get all network packet even which destined toi the others vservers [17:27] am i right ? [17:28] yes [17:28] and they can change their network setup ? [17:28] i think [17:40] stubbsd (~stubbsd@217.206.216.194) joined #vserver. [17:54] AHTOH (~Anton@212.1.230.115) joined #vserver. [17:54] ensc? i need little help [18:04] monrad (~monrad@213083190243.sonofon.dk) joined #vserver. [18:15] Nick change: Bertl_zZ -> Bertl [18:29] _Axu_: bertl is here now [18:29] hey Bertl [18:29] hi Jon! [18:40] stubbsd (~stubbsd@217.206.216.194) left irc: Ping timeout: 480 seconds [18:48] [root@vserver:dns /]/etc/init.d/named start [18:48]

Starting named: named: capset failed: Operation not permitted [18:48] anyone plz help [18:48] he has a CAP_SYS_ADMIN [18:48] already [18:49] anton: recompile named without the linux caps option [18:50] but what is about that problem? [18:50] http://www.linux-vserver.org/index.php?page=Linux-Vserver+FAQ [18:50] G. Software compatibility [18:51] bind plays with the capability system and tries to raise it's limits before giving up the caps ... [18:51] can i turn off ulimit? [18:52] i m ready to give him all [18:52] you can also post your logon/password on yahoo ... [18:52] i want to run it now in 5 minutes? is that possible? [18:52] later will recompile [18:52] sure, start it with all caps enabled ... (remove --secure) [18:52] start whom? [18:53] the apache ... [18:53] sorry bind [18:53] but can tune somehow vserver to enable that? no way? [18:53] please read the FAQ, it's CAP_SYS_RESOURCE [18:54] is not CAP_SYS_ADMIN a full set of caps? [18:55] http://archives.linux-vserver.org/200401/0081.html [18:57] o thanks [18:58] sorry for stupidness -- but i am to tired and in a hurry today :) [18:58] you are apologized ... [19:00] zev (~zev@masya.aviaserv.com.ua) joined #vserver. [19:00] hi zev! [19:00] hi [19:01] Berti i have some questions about networking inside vserver [19:01] Berti have you time? [19:02] ok. will join later... [19:02] Bertl, once more: we have one common routing kernel table or can i have different ones for each vserver? [19:03] AHTOH you speak russian? [19:03] yes [19:03] AHTOH privet [19:03] helo [19:03] AHTOH :) [19:06] anton: there is a feature called routing tables, you can setup one for each vserver ... [19:06] zev: yep, I have time now, go ahead ... [19:06] Berti is that feature in vs1.26 ? [19:07] which feature? [19:07] Berti "routing tables" [19:07] it's in vanilla 2.4.x, not vserver realted at all ... [19:07] does that feature means i can have a default gw for each vserv? [19:07] ah :) [19:07] yes, exactly ... [19:08] http://archives.linux-vserver.org/200311/0470.html [19:08] 8-) [19:08] Berti my question is exactly about Anton asked :) [19:08] good .. [19:12] Action: zev gone home... [19:12] zev (~zev@masya.aviaserv.com.ua) left irc: Quit: Trillian (http://www.ceruleanstudios.com) [19:18] Nick change: cgone -> cdub [19:18] hi cw! [19:19] Bertl: morning [19:19] hmm, chris, should I contact anybody specific at osdl regarding the kernel compile stuff? or should I just forget about doing it with osdl? [19:20] i'm in a conf. call, let me get back with you, ok? [19:20] np [19:35] monako (~monako@ts1-a111.Perm.dial.rol.ru) joined #vserver. [19:35] hi monako! [19:38] _shur1 (~shushushu@vserver.electronicbox.net) joined #vserver. [19:38] hi shuri! [19:39] <_shur1> hi [19:40] <_shur1> Bertl i found how to collect eth0:x data from vserver and put it on mrtg [19:40] <_shur1> ipttables + a littkle script to collect data [19:40] <_shur1> seem to work fine:P [19:41] yeah, that is/was what I'm saying ... (maybe not to you ;) [19:42] <_shur1> yes to me [19:42] <_shur1> i rewrite the script to collect data... [19:44] ah, okay, maybe you could add some How-To to the wiki, with some references to the required pieces ... [19:44] <_shur1> ok [19:44] <_shur1> about the cpu limite patch [19:44] <_shur1> it look fine.. [19:45] yeah, just sent a mail, so if you are interested in testing too, we can get a working version very fast ... [19:45] <_shur1> can you addatpt it to 2.4.25?? [19:46] <_shur1> of 2.4.24 [19:46] <_shur1> or [19:46] <_shur1> i can test it today [19:46] well, I'd prefer to put it in 2.6.3, but a 2.4.25-lck version is also on my todo list ... [19:47] <_shur1> i got problem with 2.6 and vserver [19:47] what issues? [19:47] <_shur1> the util-ttol do not compil [19:47] <_shur1> i past you the error last week [19:47] hmm, ever talked to enrico? [19:47] <_shur1> didnt understand your answer like usaly:P [19:48] sorry, okay, did you tell enrico about your issues with the tools? [19:48] <_shur1> yes [19:48] AHTOH (~Anton@212.1.230.115) left irc: Read error: Connection reset by peer [19:48] and, what did he say? [19:49] <_shur1> didnt understand his answer like usaly:P [19:49] <_shur1> lol [19:49] LOL [19:49] okay, that's a good one ... [19:49] do you ahve his answer somewhere? [19:49] <_shur1> no [19:50] okay, what tool versions did you try yet, and what distro do you use? [19:50] <_shur1> debian [19:50] debian woody? [19:50] <_shur1> yes [19:50] would it be an option for you to update to sarge? [19:50] <_shur1> sarge is unstable ? [19:51] well, I don't know what sarge is, but I know, that many people reported Sarge has no issues, woody has ... [19:51] <_shur1> humm [19:52] 17:11 < ensc> _shur1: call ./configure as 'CPPFLAGS="-D__NR_vserver=273" ./configure ...' [19:52] <_shur1> what is mean! [19:53] call ./configure as 'CPPFLAGS="-D__NR_vserver=273" ./configure ...' [19:53] <_shur1> i did that [19:54] does the same error happen at ./configure? Or at make-time? [19:55] <_shur1> configure [19:55] <_shur1> i'v stop testing after that... [19:56] <_shur1> apt-get dist-upgrade [19:56] <_shur1> upgrading to sarge [19:56] <_shur1> .. [19:56] _shur1: are you using stable or alpha branch? [19:59] mhepp (~mhepp@r72s22p13.home.nbox.cz) joined #vserver. [20:02] hi mhepp! [20:02] <_shur1> alpha [20:03] Bertl: hi! [20:08] _shur1: I do not understand completely why this happens, but 'ensc_cv_value_syscall_vserver=273 ./configure ...' should solve it [20:12] monako (~monako@ts1-a111.Perm.dial.rol.ru) left irc: Ping timeout: 499 seconds [20:18] <_shur1> ok [20:18] <_shur1> will retry [20:28] ben (~ben@bengrimm-host225.dsl.visi.com) left #vserver. [20:31] paul (~irssi@195.202.59.5) joined #vserver. [20:33] Moin! [20:33] hi paul! [20:44] rs (rs@ice.aspic.com) left irc: Quit: leaving [20:51] mhepp (~mhepp@r72s22p13.home.nbox.cz) left irc: Remote host closed the connection [20:54] dazedark (~chris@82-32-130-79.cable.ubr05.hawk.blueyonder.co.uk) joined #vserver. [20:55] hmm. [20:55] hi dazedark! [20:55] I think I have upset my vserver :) [20:55] hey herbert [20:55] better than your stomach! [20:55] root@asite:/# ps aux [20:55] Error: /proc must be mounted [20:55] To mount /proc at boot you need an /etc/fstab line like: [20:55] /proc /proc proc defaults [20:55] In the meantime, mount /proc /proc -t proc [20:55] proc security! [20:55] maybe I have vproc'd something I shouldn't have [20:55] :) [20:55] but ls -l /proc works [20:56] try chcontext --ctx 100 ls -l /proc [20:57] hmm, I don't have chcontext inside the vservers [20:57] on the host of course ... what does it show? [20:57] A normal ls -l of /proc [20:57] nothing odd [20:57] nothing missing? [20:58] ooh hell [20:58] loads of things are missing [20:58] you are with stable (1.26)? [20:58] yep [20:58] okay try with vproc -e /proc/[a-z]* /proc/[a-z]*/* /proc/[a-z]*/*/* [20:58] this will basically disable the proc security, so do it just for a test ... [20:59] (won't work if you did use the hide flag) [20:59] "vproc: ioctl not supported on /proc/self/lotsofthings [20:59] " [21:00] that's okay ... [21:00] can I mail you the two /proc listings? [21:01] it still doesn't work? [21:01] cos running chcontext ... after the vproc -e shows the same things [21:01] oh no, it works now [21:01] hmm [21:01] crazy [21:02] chaosle (~yvan@bragi.fh-brandenburg.de) left irc: Ping timeout: 480 seconds [21:03] ensc: could we add a sane default setup (configurable of course) to the vserver start/stop runlevel script, or did you already do that in alpha? [21:03] (I mean for the proc permissions/visibility) [21:03] Bertl: must be written; but does not happen before the general vserver initscript [21:05] what is your timeframe for this stuff, and alpha becoming stable? [21:06] ben (ben@bengrimm-host229.dsl.visi.com) joined #vserver. [21:06] hi ben! [21:06] hi Bertl! [21:09] Action: _shur1 just adding his first how to in the wiki [21:09] so, I think I'm going to merge the vserver patch for 2.4.25 with the lck patch [21:10] hmm, you are? [21:10] at least the performance patches [21:10] unless someone is already doing it... [21:11] decided I can't wait much longer on the o(1) scheduler - and at least my preliminary patching suggests that it's not going to be terribly difficult [21:11] a few rejects, but nothing crazy [21:12] hmm, you are doing what exactly? [21:12] in my vserver application? or... [21:13] in your patching ;) [21:13] ahh.. just wanting to take a stock 2.4.25, apply the ck patch set and then apply vserver [21:14] which vserver patch? [21:14] 1.3.8 [21:14] hmm, and you probably mean lck, right? [21:14] yeah lck [21:14] (said that above.. typod below) [21:15] won't work, at least the scheduling is broken after that ... [21:15] at the very least the lckbase diff [21:15] badly broken? [21:15] well, yes, that is why there are two branches (or where at least) [21:15] O(1) needs special vserver support ... [21:16] can I get that special vserver support from the previous patch? [21:16] or do you think I'd be getting in over my head ;-) [21:17] not sure ... you could have a look at the 2.6 implementation [21:17] or... how stable is the vserver patch for 2.6? [21:17] well it is probably as stable as the 1.3.x [21:18] ah, well that's probably good enough [21:18] hmm.. 2.6 something new every day [21:22] hi Bertl! [21:22] it looks like 0.29.3 compiles for me now! [21:23] great! [21:24] you had mentioned several days ago, when I was originally trying to compile it, some features that .29 has over .28 [21:24] I just read the Changelog and the News file, but none of them ring a bell [21:24] someone mentioned that the +t on /vservers is not heredetary in .29, but I think that this would be in the kernel, not in the utils [21:25] hmm, might it be that I suggested the alpha branch? [21:25] yes, you did [21:26] 0.29.3 is pre to stable, not alpha ... [21:27] Action: micah is looking at the logs [21:28] you suggested .29.2 [21:28] might be, I do not remember what the discussion was about ;) [21:28] it might be easier to tell me what you are up to ;) [21:28] hehe [21:29] I just have it on my list to resolve this :) [21:29] I can't remember why [21:29] you had asked me to send the error from the compile to enrico and the list [21:29] I am trying to remember why you were telling me to try .29.2 [21:30] ah, I was confused [21:30] okay, it's about the +t heredity issue ... [21:30] I thought .29.2 was the alpha util branch [21:30] that was what I meant with 19:28 < Bertl> 0.29.3 is pre to stable, not alpha ... [21:30] right, I've resolved the +t heredtary with scripts :) [21:31] yeah, I understand it now :) [21:31] well ok, I will probably stay with .28 until it becomes stable, and look forward to when alpha is stable for the nice things it has :) [21:31] I thought about doing a vs1.27 with the heredity issue fixed on the kernel side ... [21:31] jcollins (~jcollins@12.37.36.141) joined #vserver. [21:32] Bertl: yeah, the heredity is easily fixed with scripts so it isn't a big deal [21:32] bertl, oh well can't do 2.6 yet - no stable drbd support guess I'll have to muddle through for a couple more months [21:32] I think that I thought the .29 stuff would have the unify things that alpha has [21:32] because I thought .29 was the alpha branch [21:32] that was my opinion too, and neither 1.3.8 nor the 2.6 versions have this problem [21:33] etienne (~eroulland@81.80.240.90) left irc: Quit: Client exiting [21:33] ben, you'd test a lck version extensively? [21:33] Can anybody point me in the right direction regarding UDP issues in the host server? I've read the mailing list about fixing sshd and still no go. [21:33] so i wont need to worry about flags when 1.4 comes out then? [21:33] Basically things like 'host ' don't work. [21:34] jcollins: which kernel patches? [21:34] talon: no! [21:34] bertl, I'd make sure it works at least, put it under high load on both up and smp systems [21:34] Bertl: vs1.24 [21:34] Bertl: on a 2.4.24 kernel [21:34] hum hum, no security issues? [21:35] Bertl: same kernel on a different box works fine. [21:35] then it must be a user configuration issue, right? [21:35] Bertl: hm, the LVM guy doesn't seem to have a solution to my VGDA problem [21:36] that sounds bad, Heinz usually knows a solution to any LVM issue :( [21:37] well, he hasn't replied [21:37] Bertl: on one server /proc/self/status says "ipv4root: 0" and on the other its "ipv4root: 0100007f/00ffffff 9f064b0a/00ffffff" [21:37] I jsut assume that means he doesn't know [21:38] jcollins: that means, that one server doesn't specify IPs and the other does specify 127.0.0.1 and 10.75.6.159 ... [21:39] Bertl: You mean in calls to chbind and such for v_*? [21:39] Bertl: and the one with "ipv4root: 0" is the working server [21:39] and we are talking about vserver and not host bindings? [21:40] jcollins: give me some details about your vserver configs, maybe in private ... [21:41] the problem is not with the vservers but the vserver host [21:45] ah okay, so you are talking about ssh to the hsot, or something like this? [21:45] Purple vs. Green [21:45] hmm, day of the tentacle? [21:45] nope, B5 [21:50] Bertl: I think what lvscan and vgscan do is read /dev/lvm and compare them to /etc/lvmtab [21:51] well, vgscan is supposed to update this table [21:51] not sure about lvscan ... [21:52] its the same with vgdisplay, it reads /dev/lvm and /etc/lvmtab and then says that they differ :p [21:52] probabaly best would be to make a backup, and scratch the entire pv [21:53] yeah :p [21:53] I wonder if it is because I have /dev/md0 and /dev/md1 which used to be under LVM but aren't anymore [21:54] hmm, 'but aren't anymore' ? [21:55] right, well let me explain [21:55] I've got /dev/sda and /dev/sdb setup as a raid mirror [21:56] Personalities : [raid0] [raid1] [21:56] read_ahead 1024 sectors [21:56] md0 : active raid1 sdb1[0] sda1[1] [21:56] 96256 blocks [2/2] [UU] [21:56] [21:56] md1 : active raid1 sdb2[0] sda2[1] [21:56] 2000000 blocks [2/2] [UU] [21:56] [21:56] md2 : active raid1 sdb4[0] sda4[1] [21:56] 5807424 blocks [2/2] [UU] [21:56] /dev/md1 and /dev/md0 are NOT under LVM: [21:56] /dev/md1 1968528 224340 1644188 13% / [21:56] /dev/md0 93207 15035 73360 18% /boot [21:56] but /dev/md2 is: [21:56] /dev/rootvg/homelv 495844 14983 455261 4% /home [21:57] /dev/rootvg/usrlv 1032088 862296 117364 89% /usr [21:57] /dev/rootvg/varlv 1032088 144224 835436 15% /var [21:57]