[00:01] caligula (~junior@adsli217.cofs.net) joined #vserver. [00:09] nathan_ (~nathan@209-6-130-26.c3-0.sbo-ubr1.sbo-ubr.ma.cable.rcn.com) joined #vserver. [01:26] Nick change: riel -> unriel [01:46] Malvo lives! [02:14] Simon (~sgarner@apollo.quattro.net.nz) joined #vserver. [02:55] ccooke (~ccooke@80.1.164.238) left irc: Ping timeout: 499 seconds [02:59] ccooke (~ccooke@80.1.164.238) joined #vserver. [03:08] Nick change: Bertl_oO -> Bertl [03:09] hi simon! [03:19] okay, wish you all a merry xmas ... I'm going to bed now ;) [03:19] Nick change: Bertl -> Bertl_zZ [03:42] Doener (~doener@pD958883F.dip.t-dialin.net) left irc: Quit: Leaving [03:51] kestrel (~athomas@dialup51.optus.net.au) left irc: Ping timeout: 499 seconds [05:10] LL0rd (~dr@pD9507EF2.dip0.t-ipconnect.de) joined #vserver. [05:10] Linux_Lord (~dr@pD9507EF2.dip0.t-ipconnect.de) left irc: Read error: Connection reset by peer [05:10] Simon (~sgarner@apollo.quattro.net.nz) left irc: Read error: Connection reset by peer [05:10] Simon (~sgarner@apollo.quattro.net.nz) joined #vserver. [05:17] Simon (~sgarner@apollo.quattro.net.nz) left irc: Quit: so long, and thanks for all the fish [05:52] kestrel (~athomas@dialup51.optus.net.au) joined #vserver. [08:52] caligula (~junior@adsli217.cofs.net) left irc: Ping timeout: 485 seconds [09:54] zyong (cat@bb220-255-107-245.singnet.com.sg) left irc: [10:56] [1]James (~James@ip68-96-180-27.lv.lv.cox.net) joined #vserver. [10:56] <[1]James> hello all [11:29] mayday66 (alert@24-161-89-29.hvc.rr.com) joined #vserver. [11:29] mayday66 (alert@24-161-89-29.hvc.rr.com) left #vserver. [11:30] [1]James (~James@ip68-96-180-27.lv.lv.cox.net) left irc: Quit: HydraIRC -> http://www.hydrairc.com <- The dawn of a new age [11:36] say (~say@212.86.243.154) left irc: Ping timeout: 499 seconds [12:06] say (~say@212.86.243.154) joined #vserver. [13:26] Tamama (~Tamama@a62-216-20-152.adsl.cistron.nl) joined #vserver. [13:26] morning [13:27] I think i found a small buglet, or i did something wrong :) [13:28] when i have a vanilla kernel, compile this without vserver patch.. it works [13:29] apply patch, and after the compile/install thing it keeps complaining it can not mount the root device [13:29] 2.4.23/vs-1.22 [14:39] Madkiss_ (madkiss@madkiss.org) got netsplit. [14:39] surriel (~riel@riel.netop.oftc.net) left irc: Ping timeout: 499 seconds [14:40] Madkiss_ (madkiss@madkiss.org) returned to #vserver. [15:56] Doener (~doener@pD95880E5.dip.t-dialin.net) joined #vserver. [16:47] bah xmas sucks, everybody gone ;) [17:05] Nick change: Bertl_zZ -> Bertl [17:05] Tamama: not everyone ;) [17:11] hi [17:11] merry christmas everyone :) [17:11] merry xmas to you too ... [18:27] Action: virtuoso joins [18:27] hi virtuoso! [18:27] Hi, Bertl ] [18:27] s/ \]/!/ [18:28] ahh somebody who actually knows sed ;) [18:33] :) [18:51] surriel (~riel@imladris.surriel.com) joined #vserver. [18:51] merry xmas, master rik! [19:00] loger joined #vserver. [19:04] nathan_ (~nathan@209-6-130-26.c3-0.sbo-ubr1.sbo-ubr.ma.cable.rcn.com) left irc: Ping timeout: 499 seconds [19:08] merry christmas :) [19:08] lalalala [19:08] Action: Tamama kicks kernel [19:08] Action: virtuoso . o O ( connection timed out :) ) [19:08] it worked before on single IDE drive.... now that i have a hardware raid controller, it farts hard [19:09] Tamama: you have to configure it properly ... [19:09] and merry xmas to you too ... [19:09] configure it? [19:09] how can i configure it if it doesnt boot? heh [19:10] you changed the hardware, right? [19:10] and the kernel.. [19:10] the kernel i had before, including the config _worked_ [19:10] okay, so you can change back, right? [19:10] standard raid.s kernel [19:10] sure [19:10] i boot without vserver now [19:11] so you can configure the 'new' kernel ... [19:11] right.. and i had done so [19:11] make oldconfig [19:11] what is the 'working' kernel (version/etc)? [19:11] then it asks for the added stuff for vservers.. 'y', and voila [19:11] 2.4.23 [19:12] okay, you now do the following: download (or verify) the vanilla sources, then build a vanilla 2.4.23 kernel with your old config (make oldconfig) [19:12] i did [19:13] that one i am using now [19:13] ;) [19:13] really the only difference between the working version and the non-working version is the patch [19:13] okay, this works without any issues, and 'finds' the root device, right? [19:13] yep no problems [19:13] what compiler version? gcc? [19:14] root@powerhouse:/usr/src# gcc --version [19:14] gcc (GCC) 3.2.2 [19:15] it is a bare slackware install [19:15] do you ahve a problem installing gcc 2.95 or 3.3.2? [19:15] LL0rd (~dr@pD9507EF2.dip0.t-ipconnect.de) left irc: Quit: fractal 1.0.9b60 - Beware of programmers who carry screwdrivers. - http://fractalscript.com [19:15] I would prefer the former ... [19:17] well, i had vservers on exactly the same setup.. working, before.. except for the 3ware raid controller.. [19:18] now it needs to boot of off scsi and *pooop* [19:18] well it actually boots... [19:18] and I had big troubles with gcc 3.2.x ;) [19:18] but when it tries to mount, it farts [19:18] but it might be that you 'just' need to rebuild the initial ramdisk ... [19:19] are the drivers for 3ware raid compiled into the kernel or as modules? [19:19] kernel [19:19] since it boots from it [19:19] architexture is x86 ? [19:20] p2-350, yes [19:20] intel 440BX chipset [19:20] okay, error message on the 'failing' boot? [19:20] best would be the complete boot log (dmesg) [19:20] it can't _write_ the bootlog [19:21] sice it cant mount the file system [19:21] the only dmesgs i have are from the correct boots [19:21] serial connection or net console would help ... [19:21] true... but is absent [19:22] then you'll have to copy the last page per hand :( [19:23] fbc (~fbc@ppp-49-135.26-151.libero.it) joined #vserver. [19:23] bah [19:23] Bertl: are you there? [19:23] the monitor isnt even in this room lol [19:23] start at the end, and put it somewhere where I can read it ... [19:23] fbc: yup, I'm there ;) [19:24] Bertl: could you please send me by mail you paypal address? I forgot to save it [19:24] http://www.13thfloor.at/vserver/donate/ [19:24] ah, good! [19:25] so I conclude you like vserver, right? [19:25] No problems with vserver/ctxquota till now [19:25] i'll do the manual copy to paper and then type it on a webpage routine in a bit.. when i finish dinne r:) [19:25] okay, cu then ... [19:26] i'm almost done though hehe [19:26] I've been very busy these days, but my vservers didn't show a single problem... I like when I start something and it just works without monitoring it every minute :) [19:27] I'm glad to get some 'positive' feedback, usually you 'only' get the bugs as developer (no I'm not complaining ;) [19:28] lol [19:28] guilty as charged :D [19:28] Now, I've got my christmas question :) I'm about to test it mysqlf, but a word of advice could save me some work... if I issue iptables commands from inside a vserver, do they work? If so, could they influence other vservers addresses? [19:28] s/mysqlf/myself :) [19:28] but when you have a look at the statistics, it seems that more than 140 people are using vservers, and only a few complain ;) [19:29] and i dont even complain :P [19:29] fbc: no iptables are not allowed from inside a vserver ... not yet ;) [19:29] yet? [19:29] nooooooooooo [19:29] never! [19:29] :) [19:30] they will be 'allowed' and 'safe' when we switch to a virtual network interface ... [19:30] fbc: good to know [19:30] but what you can do easily, is setup a routing table per vserver and add/remove the entries from the host [19:32] I'm already doing something similar with ipac-ng for monitoring bandwidth usage of vservers, I was just worried if some vserver "delegate" could mess with iptables [19:33] nope, if you do not give 'dangerous' capabilities like CAP_NET_ADMIN to a vserver, it's safe ... [19:34] good [19:35] uhm, I'm reading the "donate" page, does someone of you need free hosting space? [19:36] well, I guess hosting space is the only thing we have plenty of at the moment ;) [19:36] but thanks for the offer ... [19:46] http://www.vangog.net/panic.txt [19:49] the lines where it went poop [19:50] Nick change: fbc -> fbc_dinner [19:51] hm i just had a thought.. make the kernel name smaller.. [19:51] heh [19:51] although i dont see what that should matter [19:51] Tamama: hmm, not much, especially why is kmod called when the driver is compiled in? [19:51] beats me [19:51] okay try to get at few more lines before the kmod ... [19:51] I'm back in 10-20mins ... [19:51] Nick change: Bertl -> Bertl_oO [19:52] k [19:54] fbc_ (~fbc@ppp-29-191.26-151.libero.it) joined #vserver. [19:59] Nick change: fbc_ -> fbc [20:01] fbc_dinner (~fbc@ppp-49-135.26-151.libero.it) left irc: Ping timeout: 485 seconds [20:09] Right. Time to visit home. [20:09] ccooke (~ccooke@80.1.164.238) left #vserver. [20:30] Nick change: Bertl_oO -> Bertl [20:30] okay, I'm back ... [20:33] Bertl: I'm looking at the newvserver script to see if it's possible to create a Mandrake vserver [20:33] happy holidays :) [20:33] hm heh [20:33] What holidays? :) [20:33] fbc, yup, it is possible, there should be a minimum mandrake list, but I guess it's somewhat outdated ... [20:33] it works now.. i had forgotten to actually copy the build vserver kernel that was from standard config.. it was a stripped config that didnt work (still weird) [20:34] Tamama: glad that it isn't vserver related *G* ... [20:34] i still think it's fishy! ;) [20:36] hmm, so you expect /bin/bash to boot correctly, when copied in /boot/vmlinuz, right? ;) [20:36] nah, but tcsh should! [20:36] ;) [20:37] and i dont copy over vmlinuz.. heh [20:37] hmm, right, for tcsh this might be true, at least the size suggests this *smile* [20:38] right [20:38] everybody knows the B in basm stands for big... so it wont fit in a boot! [20:38] fbc: many thanks for your donation ... enjoy your vserver ;) [20:38] many thanks for vserver [20:40] hm ok now that i have this to work.. i can just apply more patches like the quota's.. or is that already in 1.22? [20:40] quota is addon for vs1.22 ... [20:41] but remember, you need static contexts for quota/tagxid patches to work .. [20:41] ok. In preperation i already made a /vserver partition/mount :) [20:41] static contexts? [20:41] S_CONTEXT= [20:42] in the vserver config? [20:42] yup .. [20:42] oh i thought maybe it needed another patch ;) [20:42] tanjix (ViRu_@pD904A070.dip.t-dialin.net) joined #vserver. [20:43] hi [20:43] hi tanjix! [20:43] i have used the newbiew howto to setup vservers - works fine as well [20:43] just one question: can i define for each vserver a disk space amount of e.g. 4 GB ? [20:43] Bertl: other than ulimit, is there some way for limiting vserver memory usage? IIRC a similar feature is only found in a commercial spawn of an old vserver version [20:44] tanjix: yes you can limit the disk space in two ways ... [20:44] a) by using lvm logical volumes of 4GB size [20:44] b) by using the tagxid/disk limit patches [20:45] fbc: there is an experimental memory limit patch [20:45] do they work for the hwoto on the page ? [20:45] tanjix: the patches work for vserver 1.22 but I guess the howto doesn't explain how -to use them yet ... [20:45] Bertl: is the patch experimental for the stable version or for the developement version? [20:46] can you tell me how to use them ? [20:46] fbc: there should be a version for the stable somewhere ... [20:46] tanjix: sure I can ... do you want to take this path? [20:46] what do you mean ? [20:46] the lvm option is probably easier for a beginner ... [20:47] then we should use that :) as it does what i want it's ok :)0 [20:47] (I can explain that one too ;) [20:47] tanjix: I'm about to post a slighly modified version of the vserver script with support for disk space limiting in the conf files in the next few days on the ML [20:48] first explain the way i can use now to test that please :) [20:48] okay, you know lvm? [20:48] no [20:48] hmm, good, lvm means logical volume manager .. [20:48] same here :) [20:49] i see lvm is compiled in the kernel :) [20:49] that is smething included in 2.4.x since a long time, now it's somewhat depreciated in 2.6 (replaced by dm device mapper) [20:49] doh [20:49] i have 2.4.23 [20:49] the patched version of that kernel [20:49] but that doesn't mean that it doesn't work in 2.4.x [20:49] (howto) [20:50] why do they include it in 2.4 just to remove it in 2.6 again [20:50] makes no sense [20:50] well, the kernel policy is , hmm, complicated ... ;) [20:50] I would almost think 'absent' [20:50] lol [20:51] but I guess we talk about that later, if you wish ... [20:51] sure LVM is still patchable on 2.6? [20:51] or is dm 100% backwards compatible woth LVM? [20:51] okay, as usual, I put up a patch set for vserver kernels ... [20:51] Tamama: it's backward compatible, somehow/somewhat ... [20:51] right.. not then :) [20:52] the server patchset for 2.4.23vs1.22 is at http://vserver.13thfloor.at/Stuff/patches-2.4.23vs1.22/ [20:52] this includes some stuff used by myself .. [20:53] Nick change: surriel -> riel [20:53] the 03_* patch updates lvm to the most recent version (eliminating a few bugs) [20:53] ok i downloaded that [20:53] the 04_* version adds device mapper support for 2.4, but that is something different ... [20:54] do i need that ? [20:54] tanjix: you can safely apply all patches, they should not hurt ... [20:54] ok [20:54] i'll d/l them all [20:54] you get some new features, like storing the kernel config in /proc/config.bz2 ... [20:55] ok i downloaded them all now [20:56] how are those patches now applied to the kernel src [20:56] sorry, but i'm a noob when i have to work with kernels :) [20:56] cat patch | patch -p1 (if i remember correctly) [20:56] simple, you put the patches in a directory in /usr/src/ for example /usr/src/patches-2.4.23vs1.22 [20:57] hm.. s/cat patch/cat patch_file_name/; [20:57] basically you unpack your vanilla kernel in the same /usr/src/ directory [20:57] as linux-2.4.23 [20:58] then you make a linked copy like this: cp -la linux-2.4.23 linux-2.4.23vs1.22 [20:58] and change into the new directory with: cd linux-2.4.23vs1.22 [20:58] stop [20:58] what vanilla kernel [20:59] vanilla means the unmodified kernel from kernel.org [20:59] ok :) [20:59] you downloaded that, right? [20:59] yes [20:59] as bz2? [20:59] the kernel? or the patches [20:59] the ekrnel, yes [21:00] as tar gz [21:00] okay, then you unpack it with: tar xzf linux-2.4.23.tar.gz [21:00] alreadx done [21:00] -x+y [21:00] okay so you have a /usr/src/linux-2.4.23 [21:00] right? [21:00] yes [21:00] cd /usr/src [21:01] cp -la linux-2.4.23 linux-2.4.23vs1.22 [21:01] k [21:01] you now have a shallow copy of that tree ... [21:01] yes [21:01] this means, the files are actually hard links [21:01] and it uses only a fraction of the size the kernel has ... [21:01] hm [21:02] nice [21:02] now I assume that the patches are in patches-2.4.23vs1.22 in /usr/src, right? [21:02] but that will overwrite the original sources after you patch, right? [21:02] Bertle: yes, they are there [21:02] Tamama: no, the way 'patch' works, prohibits that ... [21:02] hm.. i dont know how patch works then :) [21:03] carry on, carry on ;) [21:03] i'll bug you later lol [21:03] cd linux-2.4.23vs1.22 [21:03] you are now inside the 'new' kernel tree ... [21:03] you downloaded the patches as bz2 I assume ... [21:03] yes [21:04] okay, here is a nice trick, which doesn't care what compression is used ... [21:04] but first we check if the patches are there and correct ... [21:05] for n in ../patches-2.4.23vs1.22/*; do md5sum $n; done [21:05] [root@plain src]# for n in ../patches-2.4.23vs1.22/*; do md5sum $n; done [21:05] md5sum: ../patches-2.4.23vs1.22/*: No such file or directory [21:06] well, in this case, you lied to me ;) [21:06] lol [21:06] patches-2.4.23vs1.22 is m< folder [21:06] where the patches are in [21:07] [root@plain src]# ls patches-2.4.23vs1.22 [21:07] 01_kconfig-2.4.23.diff.bz2 06_1_floppy_fallback.diff.bz2 [21:07] 02_devfs-v199.17.diff.bz2 10_2.4.23-vs1.22.diff.bz2 [21:07] 03_lvm-1.0.8-2.4.23.diff.bz2 11_i2c-2.8.2.diff.bz2 [21:07] 04_device_mapper-2.4.23.diff.bz2 12_lm_sensors-2.8.2.diff.bz2 [21:07] 05_zlib_1.1.4_updates.diff.bz2 [21:07] ah okay, you skipped the cd linux-2.4.23vs1.22 then [21:07] i see, sry :) [21:07] bad tanjix, bad! tanjix :) [21:07] Action: Tamama hides ;) [21:08] no prob Tamama :) [21:08] Bertl: check done [21:08] hmm, I guess I know who volunteered to write a howto about that, what do you think? [21:09] okay, here are the md5 sums ... [21:09] 46a53d765eafa7203bc34a507973e925 01_kconfig-2.4.23.diff.bz2 [21:09] 03cb87239d263cac6a49b41a4885352d 02_devfs-v199.17.diff.bz2 [21:09] 3c2155439762fbd817476bf5255995f1 03_lvm-1.0.8-2.4.23.diff.bz2 [21:09] 060e1a0d44e9b6318dbfb4ef5a1897de 04_device_mapper-2.4.23.diff.bz2 [21:09] 0ca4e26d505133845454848da9960f8d 05_zlib_1.1.4_updates.diff.bz2 [21:09] dd367927e0b971158c52c8334f11e20b 06_1_floppy_fallback.diff.bz2 [21:09] 71cf563fbb2ed71eaa315d23a5ecc3d4 10_2.4.23-vs1.22.diff.bz2 [21:09] 80508d9a6e8e40f092936ea8c0db1636 11_i2c-2.8.2.diff.bz2 [21:09] c27f956512751893c81a6b05c7948506 12_lm_sensors-2.8.2.diff.bz2 [21:09] they are correct [21:10] perfect, now we use almost the same line to patch the sources ... [21:10] for n in ../patches-2.4.23vs1.22/*; do less $n | patch -p1 ; done [21:10] done [21:11] you got a list of files patched, right? [21:11] yes [21:12] if, for any reason, you find something which reads like 'failed' or 'fuzz' then you have a problem ... [21:12] nothing to see [21:12] 'offset' and 'succeded' are okay ... [21:12] fine .. [21:12] serving (~serving@213.186.189.29) left irc: Ping timeout: 499 seconds [21:13] now we need a kernel config, do you already have one? [21:13] i have one from the howto [21:13] ftp://ftp.solucorp.qc.ca/pub/vserver/config-2.4.22 [21:13] hmm, okay, so be it ... [21:14] but let me say that one of the issues with 'generic' kernel configs is, they do not support your system the way the could, and always add some bloat you do not need ... [21:15] those are two reasons, why I don't provide such configs ... [21:15] i see [21:15] but you can start with that one, as it actually works (right?) [21:15] it works, yes [21:16] okay, then copy that one into /usr/src/linux-2.4.23vs1.22 as .config [21:16] and do: make oldconfig [21:16] done [21:16] that will ask you everything added since this config was made ... [21:17] done [21:17] look at each question, if you do not understand what it means, take the default ... [21:17] okay, now we have to make sure that lvm and vroot is activated ... [21:17] i agreed to all wirth "NEW" was that ok [21:18] actually no, but you can do it again, just copy over the config ... [21:18] and what should i do then when running make oldconfig [21:18] 19:18 < Bertl> look at each question, if you do not understand what it means, [21:18] take the default ... [21:18] and the default is ? [21:19] that which is selected (upper case) and will be used, when you press enter ... [21:19] the word in brackets ? [21:19] yes that was NEW in everything :) [21:19] as i saw that right [21:19] NEW isn't an option, the options are Y/N/M [21:20] Provide kernel config via /proc (CONFIG_PROC_CONFIG) [N/y/m/?] (NEW) [21:20] e.g. [21:20] means N is the default, if you press enter [21:20] ok :) [21:20] in that case, you want a Y here, for example ... [21:20] o.O [21:20] (because you want the config stored in /proc/config.bz2, right? [21:21] i think yes but i dont understand a lot from kernel and its options [21:21] that is why I explain it to you ;) [21:22] yea but you said thing i dont know i should take the default :) i dont know any :) [21:22] okay, take all defaults, they won't hurt ... [21:22] ok, done [21:22] now we have to do 'make menuconfig' and select some specific features ... [21:22] bah vserver-0.29 needs linunxconf (yes i know it was already discussed but i didnt install it ;)) [21:22] in it ... [21:23] Tamama: if you don't install the admin package it doesn't need linuxconf ... [21:23] tanjix: look for lvm/raid ... [21:24] is it in the main menu or in any submenu [21:24] in the main menu (RAID and LVM), enable multiple devices support by pressing Y or 'space' and [21:25] select Logical volume manager (LVM) support [21:25] i dont have that menu [21:25] what menus do you have? [21:25] Code maturity level options ---> ¦ ¦ [21:25] ¦ ¦ Loadable module support ---> ¦ ¦ [21:25] ¦ ¦ Processor type and features ---> ¦ ¦ [21:25] ¦ ¦ General setup ---> ¦ ¦ [21:25] ¦ ¦ Memory Technology Devices (MTD) ---> ¦ ¦ [21:25] ¦ ¦ Parallel port support ---> ¦ ¦ [21:25] ¦ ¦ Plug and Play configuration ---> ¦ ¦ [21:25] ¦ ¦ Block devices ---> ¦ ¦ [21:25] ¦ ¦ Multi-device support (RAID and LVM) ---> ¦ ¦ [21:25] ¦ ¦ Networking options ---> [21:25] halt [21:25] there it was [21:25] lol [21:25] Multi-device support (RAID and LVM) --- [21:25] ahhhhh [21:26] ok i think i need glasses :) [21:26] well the menuconfig structure needs some time to get used to ... [21:26] Logical volume manager (LVM) support [21:26] a special feature are appearing and disappearing entries ... [21:27] so you don't want a module 'M' there, you want a '*' there ... [21:27] done [21:27] just press 'Y' or 'space' again ... [21:27] okay ... [21:27] yea [21:27] now the Block devices ---> [21:27] there we have to enable the vroot device ... [21:28] if you want to know what an option does, there is a help ... [21:28] < > Virtual Root device support ? [21:28] yes, you can configure that as module ... [21:29] done [21:29] do you ahve 'lspci' installed? [21:29] i think it is [21:29] okay show me the output, and I tell you what devices you'll need ... [21:29] [root@plain linux-2.4.23vs1.22]# lspci [21:29] 00:00.0 Host bridge: Intel Corp. 82845G/GL [Brookdale-G] Chipset Host Bridge (rev 03) [21:29] 00:02.0 VGA compatible controller: Intel Corp. 82845G/GL [Brookdale-G] Chipset Integrated Graphics Device (rev 03) [21:29] 00:1d.0 USB Controller: Intel Corp. 82801DB USB (Hub #1) (rev 02) [21:29] 00:1d.1 USB Controller: Intel Corp. 82801DB USB (Hub #2) (rev 02) [21:29] 00:1d.2 USB Controller: Intel Corp. 82801DB USB (Hub #3) (rev 02) [21:29] 00:1d.7 USB Controller: Intel Corp. 82801DB USB EHCI Controller (rev 02) [21:29] 00:1e.0 PCI bridge: Intel Corp. 82801BA/CA/DB PCI Bridge (rev 82) [21:29] 00:1f.0 ISA bridge: Intel Corp. 82801DB ISA Bridge (LPC) (rev 02) [21:29] 00:1f.1 IDE interface: Intel Corp. 82801DB ICH4 IDE (rev 02) [21:29] 00:1f.3 SMBus: Intel Corp. 82801DB SMBus (rev 02) [21:29] 00:1f.5 Multimedia audio controller: Intel Corp. 82801DB AC'97 Audio (rev 02) [21:29] 01:0d.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10) [21:30] okay, in network device support ... [21:30] fbc (~fbc@ppp-29-191.26-151.libero.it) left irc: Quit: I'm hungry [21:31] Ethernet (10 or 100Mbit) ---> [21:31] you leave the Realtek selected, but you can disable everything else ... [21:32] RealTek RTL-8139 C+ PCI Fast Ethernet Adapter support (EXPERI¦ ¦ [21:32] ¦ ¦ RealTek RTL-8139 PCI Fast Ethernet Adapter support ¦ ¦ [21:32] ¦ ¦[ ] Use PIO instead of MMIO ¦ ¦ [21:32] ¦ ¦[ ] Support for uncommon RTL-8139 rev. K (automatic channel equ¦ ¦ [21:32] ¦ ¦[*] Support for older RTL-8129/8130 boards ¦ ¦ [21:32] ¦ ¦[ ] Use older RX-reset method ¦ ¦ [21:33] that is okay ... but you won't need the old support .. but it should not hurt ... [21:33] in Character Device ... [21:33] what do you mean with keep real selected but disable everything else [21:34] RealTek is the network card ... there are probably others selected ... [21:34] when i deselect a few i dont have realtek anymore [21:34] you probably deselect the EISA/PCI/ support ;) [21:34] *ooops* :) [21:34] love those disappearing items [21:35] you're right :) [21:35] hm [21:35] i did a: vserver name build .... it now copies 2 kernels to it ... *rolls eyes* [21:36] this isn't util-vserver, right? [21:36] inside character device now [21:36] Bertl: ne? it is [21:36] there you can disable AGP/etc for non Intel stuff ... [21:37] enable 'Support for console on serial port' [21:37] and disable all 'dri' options ... [21:37] what is 'dri' ? direct ram interface or something? [21:37] Tamama: this is util-vserver? which version? [21:37] dri = Direct Rendering Interface ;) [21:38] 0.26, the one that is listed on the vs1.22 stable page [21:38] (something every vserver needs) [21:38] oh that.. i disabled it [21:38] hmm, and it copies ther kernels? [21:38] they are in usr/src [21:38] and it copies /usr/src [21:38] Action: Tamama rolls eyes [21:39] mm, okay, that might be true ... [21:39] Support for console on serial port i dont have that have looked 3 times [21:39] okay, you ahve Standard/generic (8250/16550 and compatible UARTs) serial support? [21:40] enable that one, the other will appear ;) [21:40] it is a M in front of that [21:40] but no serial port option [21:41] change the M to a '*' [21:41] ahh [21:41] now :) [21:41] now to 'Filesystems' [21:41] stop [21:41] (main menu) [21:42] yup, what's up? [21:42] stil looking for dri options [21:42] Direct Rendering Manager (XFree86 DRI support) ---> [21:42] ? [21:42] yup that ... disable everything if anything is enabled ... [21:43] [ ] Direct Rendering Manager (XFree86 DRI support) [21:43] everything under that disappereaed [21:43] okay, that is perfect ... [21:43] ok.. inside filesyxstems now [21:43] would you like to ahve quota support inside the vservers? [21:43] yes [21:44] so vserver root can assign quota for the other users ... [21:44] okay, then select quota support and VFS v0 quota format support [21:44] as '*' ? [21:45] or M [21:45] as you prefer ... [21:45] whats the difference [21:45] '*' means inside the monolithic kernel .. 'M' is a module in /lib/ hm what option is better ? [21:46] they are equivalent in that case ... [21:46] i used M [21:46] on VFS v0... [21:47] oh! [21:47] okay ... deselect filesystems you wont use ... [21:47] that quota support eluded me.. i thought make oldconfig was supposed to tag that for me :/ [21:48] oldconfig only shows 'new' config options ... [21:48] ok i disabled them all i dont use [21:48] and this isnt new? [21:48] heh [21:48] nope, it's in 2.4.22 too ... [21:48] vserver didn't change anything there ;) [21:49] what does it do? [21:49] quota? or the VFS v0 stuff? [21:49] VFS v0 [21:49] i know what quota does heh [21:49] ext3, compressed rom, virtual memory..., iso 9660, /proc, /dev/pts are enabled [21:49] tanjix: looks good, ext2 should be enabled too ... [21:50] where is that [21:50] just for the case that ext3 fails, and you want to repair something ... [21:50] Second extended fs support [21:50] that is enapled with * [21:50] forgot to list it :) [21:51] Tamama: Honza changed the format of quota twice ... and the 'newest' quota scheme is called version 0 ;) [21:51] oh goodie [21:51] to keep thinks easy huh [21:51] the oldest is called v1, and the second was called v2 ;) [21:52] i guess he is starting over lol [21:52] okay tanjix, now to the ernel Hacking menu ;) [21:53] Kernel Hacking I mean ... [21:53] ok [21:53] hm what to enable there then? [21:54] Kernel debugging, Magic SysRq key [21:54] [ ] Kernel debugging ¦ ¦ [21:54] ¦ ¦ (0) Kernel messages buffer length shift (0 = default) [21:54] same here [21:54] enabled .. [21:56] ahh, I forgot, in Filesystem, just below /proc ... there should be the kernel config option ... [21:57] Provide kernel config via /proc [21:57] why would you want that [21:57] ? [21:57] enabled that with M [21:57] because you can get the config of the currently running kernel by zcat /proc/config.bz2 [21:58] you'll never ever have to search for the config again ;) [21:58] hm i never do anyway [21:58] its on my cdrom :) [21:58] okay, now exit with saving the config ... [21:59] done [22:00] Tamama: it might not sound that useful with 1 kernel, but when you have 20 versions of 2.4.23, you'll be happy to know the config of the running one, right? [22:01] maybe :) [22:01] okay, now we build the dependancies, by: make dep >../Dep.log [22:01] running ... [22:02] you'll get a message about md5 sums ... [22:03] it flooded with errors [22:03] hmm, lets see a few lines ... [22:03] [root@plain linux-2.4.23vs1.22]# make dep >../Dep.log [22:03] acsi.c:74:25: asm/atarihw.h: No such file or directory [22:03] acsi.c:75:27: asm/atariints.h: No such file or directory [22:03] acsi.c:76:28: asm/atari_acsi.h: No such file or directory [22:03] acsi.c:77:29: asm/atari_stdma.h: No such file or directory [22:03] acsi.c:78:29: asm/atari_stram.h: No such file or directory [22:03] au1000_gpio.c:41:24: asm/au1000.h: No such file or directory [22:03] au1000_gpio.c:42:29: asm/au1000_gpio.h: No such file or directory [22:03] hp_psaux.c:41:26: asm/hardware.h: No such file or directory [22:03] hp_psaux.c:43:21: asm/gsc.h: No such file or directory [22:03] fetchop.c:59:25: asm/machvec.h: No such file or directory [22:03] fetchop.c:60:24: asm/sn/sgi.h: No such file or directory [22:03] that's okay ... [22:04] will go away after the first build ... [22:04] nothing to worry about ... [22:04] after that we do: make bzImage modules >../Build.log [22:06] ok running [22:06] here too [22:06] it'll be done in 1 hour :P [22:06] 7h10e7h10e [22:07] i was looking at the quota patches.. but the site only shows vs-1.20 [22:07] fbc (~fbc@ppp-105-136.26-151.libero.it) joined #vserver. [22:08] Bertl: is it possible to change a security context for a device with chctx? [22:08] Bertl: or, in other words, recreate the device node with a particular context id [22:10] it again brings up some errors [22:10] warnings [22:11] amd74xx.c:389: warning: `ata66_amd74xx' defined but not used [22:11] just ignore those :) [22:14] {standard input}:1896: Warning: indirect lcall without `*' [22:14] {standard input}:1987: Warning: indirect lcall without `*' [22:14] ignore those [22:14] and i can see your processor is faster than mine ;) [22:14] 7h10e7h10e [22:15] Action: Tamama kicks the P2-350 [22:15] lol [22:15] P4 2,4 Ghz :) [22:15] that helps [22:15] i have one of those.. but the mobo is fried :( [22:16] so preparing it on the 350.. gah so slow [22:16] bertl will the running vservers defunction when using that new kernel ? [22:19] defunction? [22:19] oh i just got to the indirect lcall warnings :) [22:20] that they dont work anymore [22:20] g its silence on most irc channels [22:20] tanjix: arent they supposed to start working with that kernel? [22:20] what could that be *g [22:21] i dont know that why i ask [22:21] you dont have vservers now? [22:21] i created two [22:21] with the howto on the page and that kernel [22:22] and in the config you put 'ON_BOOT=yes' or soemthing for the vservers? [22:22] yes [22:22] then they should just start again [22:22] ok [22:23] okay, I'm back ... give me few seconds to catch up ... [22:24] dont hurry, still compiling :) [22:24] fbc: device nodes do not use this information ... [22:25] tanjix: warnings are okay, you normally don't see them, as they are shaded by the amount of information scrolling by ... [22:25] Bertl: are you sure? Having a different security context than 0 and that of a particular vserver, I have problem with permissions [22:26] hmm, let me see ... [22:27] tanjix: existing vservers will still work ... but we have to 'modify' your partition table too ;) [22:27] ok [22:27] read 'destroy and screw over from the left nostril' [22:27] :) [22:27] the host system itself will not be affected in functionality ? [22:27] nope, if you do everything correctly, nothing will be affected ;) [22:27] applicom.c:532:2: warning: #warning "Je suis stupide. DW. - copy*user in cli" [22:27] i love that one [22:28] good :) [22:28] fbc: what issues do you encounter? [22:28] ake[3]: *** [i2c-matroxfb.o] Error 1 [22:28] make[2]: *** [_modsubdir_matrox] Error 2 [22:28] make[1]: *** [_modsubdir_video] Error 2 [22:28] make: *** [_mod_drivers] Error 2 [22:28] that looks fatal ... [22:29] 7:10( [22:29] Bertl: I encountered that issue when beginning with vserver for /dev/tty, I thought it was it again but I isolated the problem now [22:29] but don't worry, probably only a configuration issue ... [22:29] where to start looking for the prob ? [22:29] Action: Tamama imaginary sees Bertl being pulled from left to right and back again [22:29] tanjix: do make menuconfig again ... [22:30] Tamama: I'm used to that ... [22:30] usually I'm able to handle 4-6 people simultaniously ;) [22:31] fbc: hmm, after looking at the code, it _is_ possible to have device nodes with ctx info ... [22:32] fbc: but I'm not sure that chctx can modify them ... [22:33] hm, pkgtool doesnt seem to work correctly in a vserver [22:33] what is it supposed to do? [22:33] add/remove packages [22:33] it can modify for sure things like /dev/shm and /dev/null, but I don't know if these qualify as "special" dev files :) [22:34] /dev/null does ... [22:35] tanjix: in the menu config? [22:35] yes [22:35] okay Character devices ... [22:35] i2c ... [22:36] anything enabled there? [22:38] yes [22:38] I2C support ¦ ¦ [22:38] ¦ ¦ I2C bit-banging interfaces ¦ ¦ [22:38] ¦ ¦ Philips style parallel port adapter ¦ ¦ [22:38] ¦ ¦ ELV adapter ¦ ¦ [22:38] ¦ ¦ Velleman K9000 adapter ¦ ¦ [22:38] ¦ ¦ < > NatSemi SCx200 I2C using GPIO pins ¦ ¦ [22:38] ¦ ¦ < > NatSemi SCx200 ACCESS.bus ¦ ¦ [22:38] ¦ ¦ I2C PCF 8584 interfaces ¦ ¦ [22:38] ¦ ¦ Elektor ISA card ¦ ¦ [22:38] ¦ ¦ [ ] I2C mainboard interfaces [22:38] I2C device interface ¦ ¦ [22:38] ¦ ¦ I2C /proc interface (required for hardware sensors) [22:38] i2c is enabled on the matroxfb for ddc [22:38] okay, disable the matrox stuff ... [22:39] merry xmas, dan! [22:39] :) [22:39] i just disabled i2c :) [22:39] matrox staff ? [22:40] is this all you see in i2c? [22:40] yes [22:40] okay, lets go for lmsensors ... [22:40] check there ... [22:40] obj-$(CONFIG_FB_MATROX_I2C) += i2c-matroxfb.o [22:41] hardware sensors support ? [22:41] yes ... [22:41] [ ] Hardware sensors support ¦ ¦ [22:41] ¦ ¦ [22:41] disabled ... [22:41] dan, where is the framebuffer stuff? [22:41] Frame-buffer support [22:41] top level? [22:42] hmm [22:42] let me run menuconfig and see :) [22:42] ah [22:42] Console drivers->frame-buffer support [22:42] Matrox I2C support [22:42] okay thanks, tanjix, look there ... [22:43] tanjix: do you have a matrox video card? [22:43] no [22:44] ah, nm then :) [22:44] dan, that is the advantage of using jacks generic .config ;) [22:44] hehe [22:45] not that using my 'generic' config, if there was such thing ;) would be any better ... [22:45] Console drivers->frame-buffer support: [22:45] Matrox acceleration (EXPERIMENTAL [22:45] generic config: _everything_ enabled hehehe *bloat* [22:45] yep, disable everything there ... I guess you don't need fb support anyway ... [22:45] disable ALL ? [22:45] do you use fb support now? [22:45] [ ] Support for frame buffer devices (EXPERIMENTAL) [22:46] check with lsmod ... [22:46] i dont know [22:46] [root@plain linux-2.4.23vs1.22]# lsmod [22:46] Module Size Used by Not tainted [22:46] isofs 20044 1 (autoclean) [22:46] loop 11032 3 (autoclean) [22:46] autofs 12244 0 (autoclean) (unused) [22:46] sorry for answering _before_ your question [22:46] apm 11976 1 [22:46] 8139too 15432 1 [22:46] mii 3720 0 [8139too] [22:46] crc32 3680 0 [8139too] [22:46] serial 51748 1 (autoclean) [22:46] keybdev 2720 0 (unused) [22:46] mousedev 5236 0 (unused) [22:46] hid 23204 0 (unused) [22:46] input 5568 0 [keybdev mousedev hid] [22:46] usb-uhci 24652 0 (unused) [22:46] usbcore 73088 1 [hid usb-uhci] [22:46] rtc 7964 0 (autoclean) [22:46] is your screen regular text 80x25? :) [22:47] no i only have ssh access :) [22:47] well, no fb module there, any options with '*' instead of 'M' in that menu? [22:47] tanjix: ahaha, yeah no fb needed :) [22:47] so completely disable it ? [22:47] yup [22:48] okay, now again save/exit .. and start the make bzImage modules >../Build.log again [22:48] ok [22:48] running .. [22:50] on my machine too ;) [22:51] make[2]: Circular /usr/src/linux-2.4.23vs1.22/include/linux/netfilter_ipv4/ip_conntrack_helper.h <- /usr/src/linux-2.4.23vs1.22/include/linux/netfilter_ipv4/ip_conntrack.h dependency dropped. [22:51] but still running [22:51] yeah, a 'clean' kernel compile, without any warnings, wasn't seen since ... hmm linux-0.1 I guess ... ;) [22:52] 7h10e7h10e [22:54] ok done [22:54] the last lines: [22:54] {standard input}: Assembler messages: [22:54] {standard input}:251: Warning: indirect lcall without `*' [22:54] {standard input}:347: Warning: indirect lcall without `*' [22:54] [root@plain linux-2.4.23vs1.22]# [22:55] okay, that means kernel and modules compiled ... [22:55] now we install the modules with: make modules_install [22:55] done [22:56] perfect, now we copy ther kernel to /boot ... with: [22:57] cp -va arch/i386/boot/bzImage /boot/vmlinuz-2.4.23vs1.22 [22:57] and the same with the System.map: [22:57] ok, I'm out. happy holidays everyone! and may your new vserver be bright :) [22:57] cp -va System.map /boot/System.map-2.4.23vs1.22 [22:57] bye dan! have a nice one ... [22:58] thanks MrBowb :) [22:58] bertl: dione [22:58] done [22:59] okay, you are using grub or lilo? [22:59] lilo [22:59] or soemthing excotic like syslinux? [23:00] okay, add a config section for your new kernel ... [23:00] you know how to do that? [23:00] i think so [23:00] good, do you have any initrd setting for the current kernel? [23:01] no i dont think [23:01] just copy/paste an existing one and change the label and boot file :D [23:01] ehm stop [23:01] initrd=/boot/initrd-2.4.23-vs1.22.img [23:01] is the actual setting for the 1st vserver kernel [23:02] okay, then we use the same for the new one ... [23:02] image=/boot/vmlinuz-2.4.23vs1.22 [23:02] label=vserver1 [23:02] initrd=/boot/initrd-2.4.23-vs1.22.img [23:02] read-only [23:02] root=/dev/hda3 [23:02] this is my new section [23:02] label=vserver1 [23:02] and don't put the new kernel as default ... [23:02] and default=vserver1 [23:02] no ? [23:02] ok [23:03] let the old 'working' be default for now ... [23:03] ok changed back zo the other kernel [23:03] we can boot after, lilo -R vserver1 [23:03] ah testing the kernel :) [23:03] exactly ... [23:04] hmm [23:04] [root@plain linux-2.4.23vs1.22]# lilo -R vserver1 [23:04] Fatal: No image "vserver1" is defined [23:04] is first must run lilo ? [23:04] and then lilo -R vserve1 ? [23:04] yup ... [23:04] ok [23:04] [root@plain linux-2.4.23vs1.22]# lilo -R vserver1 [23:04] [root@plain linux-2.4.23vs1.22]# [23:04] should there happen anything ? [23:04] reboot -r now [23:04] [root@plain linux-2.4.23vs1.22]# reboot -r now [23:04] usage: reboot [-n] [-w] [-d] [-f] [-i] [-p] [23:04] -n: don't sync before rebooting the system [23:04] -w: only write a wtmp reboot record and exit. [23:04] -d: don't write a wtmp record. [23:04] -f: force halt/reboot, don't call shutdown. [23:04] -p: power down the system (if possible, otherwise reboot) [23:05] shutdown -r now :) or simply reboot [23:05] shutdown -r now ... sorry ;) [23:05] ok rebooting [23:06] i envy your faster cpu [23:06] mine is still building modules [23:06] lol [23:06] 7h10e7h10e [23:06] serving (~serving@213.186.190.157) joined #vserver. [23:06] server is up again [23:07] well, looks good so far, what do you think? [23:07] does it use the new kernel or the old ? [23:07] uname -a [23:07] [root@plain root]# uname -a [23:07] Linux vserver.xl-server.org 2.4.23-vs1.22 #2 Wed Dec 24 14:49:48 CST 2003 i686 i686 i386 GNU/Linux [23:07] check the build time ... [23:08] what's the time on that server now? [23:08] Wed Dec 24 15:09:35 CST 2003 [23:08] looks like that's the new one ... [23:08] cool [23:08] okay, now we have to modify your disk layout/partitions ... [23:09] this is for quota's inside vservers? :) [23:09] you have the existing vservers on one partition? [23:09] question: [23:09] go ahead ... [23:09] can the disk space be changed for existing vservers? or do i have to delete them [23:10] what patches do i need to download to enable the quotas in 2.4.23/vs-1.22 [23:10] with lvm, you can change the disk space by extending/reducing the lvm volume, which requires the filesystem to be resized, and therfore, the vserver has to be offline ... but it doesn't require you to recreate it ... [23:11] ok all vservers are stopped [23:11] with the other approach (my patches) you can change it on the fly, but it's a little trickier to get it right ... [23:12] what is your current disk layout? (sfdisk --dump will tell) [23:12] [root@plain root]# sfdisk --dump [23:12] # partition table of /dev/hda [23:12] unit: sectors [23:12] [root@plain root]# sfdisk --dump [23:12] # partition table of /dev/hda [23:12] unit: sectors [23:12] /dev/hda1 : start= 63, size= 208782, Id=83, bootable [23:12] /dev/hda2 : start= 208845, size= 4144770, Id=82 [23:12] /dev/hda3 : start= 4353615, size=112872690, Id=83 [23:12] /dev/hda4 : start= 0, size= 0, Id= 0 [23:12] okay, hda3 is your vserver partition? [23:12] Filesystem Size Used Avail Use% Mounted on [23:12] Last message repeated 1 time(s). [23:12] /dev/hda3 53G 3.9G 47G 8% / [23:13] aha, uhum ... okay ... [23:13] problem ? :) [23:13] hda1 = boot, hda2 = swap? [23:13] Filesystem Size Used Avail Use% Mounted on [23:13] /dev/hda3 53G 3.9G 47G 8% / [23:13] /dev/hda1 99M 17M 77M 18% /boot [23:13] none 442M 0 442M 0% /dev/shm [23:13] /root/iso/shrike-i386-disc1.iso [23:13] 638M 638M 0 100% /mnt/cdrom [23:14] that is "df -h" [23:14] okay ... that's a little trickier ... you have a bootdisk at hand? [23:14] no.. as i said i only can access that server per ssh [23:14] okay, how do/did you install it? [23:15] it was preinstalled [23:15] it's a dedicated server in a datacenter [23:15] hmm, I see, you have another machine there? [23:15] yes [23:16] in the same network? [23:16] should be [23:16] yes [23:16] but you don't ahve a serial connection between those machines, right? [23:16] right [23:17] hmm, let me medidate about that for a moment ... [23:17] sure [23:18] do you ahve a way to reboot that machine if it is stuck? [23:20] yes [23:22] okay, the only way I see right now would be the following: [23:23] configure the second machine for netbooting the first one (means nfs server and bootp/dhcpd/tftp), then compile a netboot kernel on the first ... then boot via network ... [23:23] you can then login to the netbooted amchine and modify the partitions, without destroying anything ... [23:24] it's a bad thing to have everything on / anyway ... [23:24] uhh i dont know how to do that all [23:24] well, it's christmas, I'll help you ;) [23:25] lol [23:25] but you have to document it, and publish the how-to ... [23:25] santa-Bertl [23:25] hohoho! [23:25] hm and i can't change another server fr doing that there is lot of applications that must run :) [23:26] well, enabling/using nfs isn't that problematic ... [23:28] how long does that all take.. isn't there an easier method to limit each vservers disk space :) [23:28] sure, have a dedicated vserver partition in the first place :D [23:28] sure, you can use and enable my patches, but soon you'll end up with a non functioning server ... [23:29] and completely reinstalling that server is that possible via remote ? [23:29] otherwise I would not have suggested it ;) [23:29] so that we can set the partitions again [23:29] it is a dedicated unmanaged server i assume? [23:30] tanjix: do you have a way to repartition that vserver remotely? [23:30] (with the current interface, whatever it might be?) [23:30] i cant say you that [23:30] dont know [23:30] I assume not, but you can do that, if you get the netboot working ... [23:31] i'm not a linux expert.. i know how to use but not everything hehe [23:31] can this be any server or must it be the same [23:32] what the same? [23:32] exactly the same server as that one we want to change [23:32] no, but it has to be in the 'same' network ... [23:33] i hope that it is [23:33] we'll see that ... [23:33] with a windows server we cant do anything ? [23:34] well, if you install nfs, bootp, and tftp, sure ;) [23:34] you can reformat it :D [23:34] ok i got a vserver running again.. wheee [23:35] but i cant remove any packages :( [23:35] tanjix: okay, think about it, if you want to do it, just let me know ... [23:36] bertl: how long (time) do you think will it take [23:37] hmm, about 30 mins until the netboot is up, 20mins to copy the root image ... [23:37] guess an hour or so ... [23:38] and do you have the time for that ? :) [23:38] hey, I'm not eager to do that, I've done that more than once ... and I can live without it .. but I said, I'll help you, right? [23:39] hey sorry for asking no reason to get loud was just a question [23:40] I'm not loud, when I'm loud, I SOUND LIKE THIS! [23:40] NO THAT IS NOT LOUAD#^&@^%%@$%^ THISIS @^^#@^%! [23:40] :) [23:41] tanjix: and I don't think that anybody really can anoy me so easily ... [23:45] Bertl: can you point me to the patch(es) to use to enable vserver quotas? :D [23:45] http://www.13thfloor.at/vserver/s_addons/quota/ [23:45] they state vs1.20 though [23:46] or doesnt that matter? [23:46] yep 1.21 and 1.22 where bugfixes only ... [23:46] ok.. nice to know :) [23:47] so.. how does 'patch' work? as you said it doesnt change the originals.. [23:48] it first moves the original file to .orig (or the extension you specify) ... then makes a copy and applies the patches (in that process, actually) [23:49] so if you hardcopy it, it just renames the link? [23:49] erm.. hardlink it [23:50] Persistent Context ID for files (Disabled, UID32/GID16, UID24/GID24, UID32/GID32) [23:50] hm.. [23:50] why not 16/16? :D [23:51] shuri (~ipv6@cpu183.adsl.qc.bellglobal.com) joined #vserver. [23:51] marry christmas Bertl [23:55] marry xmas to you shuri! [23:56] Tamama: 16/16 would be overkill, or what do you want to do with 4mio contexts? [23:56] Tamama: hard link means another name entry for an inode in a directory inode ... [23:57] well wouldnt 32/32 be more overkill? [23:57] 32/32 means that you do't use UID/GID at all, this is the 'unused' space in on disk inode option, just read the help ;) [23:58] shuri (~ipv6@cpu183.adsl.qc.bellglobal.com) left irc: Quit: ipv6 [23:58] well apperantly 24/24 just nibbles off a byte from the GID/UID.. [23:59] exactly, giving 16 bit for XID ... [23:59] and 32/32 pretty much is a hack [23:59] heh [00:00] --- Thu Dec 25 2003