[00:08] WSU (~Josh@ny.webpipe.net) joined #vserver. [00:08] hi WSU! [00:08] HI [00:09] is there a vserver for UML users page? [00:09] a special UML on vserver or vserver on UML page? [00:10] not that I know of ... [00:10] no, more a migration HOWTO [00:10] hmm, you want to migrate from ? to ? [00:12] I am using UML now and am quite happy with it [00:12] but I am offered a vserver account somewhere [00:12] where I have to provide a tarball of the guest machine [00:12] so I end up having to use vserver too [00:12] hmm, better stay away ;) [00:13] you could end up, converting all your uml stuff to vserver ;) [00:13] well, it might be better! [00:14] heh, intel goes for 30 stage pipeline [00:14] mids: basically a transition from UML to vserver is very similar to a transition from vmware or a 'normal' dedicated server to vserver ... [00:15] there is a howto for that, but it's not maintained AFAIK ... http://www.linux-vserver.org/index.php?page=VirtualizeHowto [00:15] thanks [00:16] if you have specific questions, just ask ... [00:16] miller7 (none@213.239.180.106) left irc: Ping timeout: 492 seconds [00:16] Bert. what was the syntax for the dump | restore you gave me yesterday? [00:16] I guess I'll go experiment first [00:16] then read docs [00:16] and then ask :) [00:16] WSU: probably dump 0sf 100000 - | restore rf - [00:17] WSU: dump 0sf 100000 - [/] | restore rf - [00:17] yes [00:17] ok [00:17] thanks [00:19] And that maintain +it attributes and hardlinks, right? [00:19] cd .. [00:19] yup [00:22] this hyphen is not needed, correct? 100000 - [/] [00:23] or, um that actually dump is to stdout, right? [00:27] yes [00:32] Action: WSU is having problems [00:33] hmm, what kind of problems? [00:33] I need to move everything from /dev/hdb1 [00:33] to /dev/hda5 [00:33] yeah, you said that some days ago ;) [00:33] /dev/hdb1 is mounted on /mnt/tmp [00:34] /dev/hda5 is /vservers [00:34] Yes, I did it then [00:34] messed it up since then [00:34] and don't have my irc logs at this pc [00:34] okay, anything important on /dev/hda5 [00:34] nope [00:34] good, umount, make a clean fs, remount [00:36] -k- [00:36] done [00:36] cd /vserver [00:36] cd /vservers [00:36] JonB (~NoSuchUse@129.142.112.33.ip.tele2adsl.dk) left irc: Quit: Leaving [00:37] dump 0sf 100000 - /dev/hdb1 | restore rf - [00:39] great. thanks [00:40] np [00:40] Doener_zZz (~doener@pD9E121A4.dip.t-dialin.net) joined #vserver. [00:40] sleepwalking? [00:41] Nick change: Doener_zZz -> Doener`` [00:41] back to my 'perl-correct' nick ;) [00:41] maybe i should spent the time to setup a bnc [00:41] ah that is perl, I always thought it was an acci-dent ;) [00:42] isn't perl? :) [00:43] mygrant (~mygrant@migo.cisco.com) joined #vserver. [00:43] hi mygrant! [00:43] hey Bertl [00:44] what brings cisco? here? [00:44] I've been looking at this vserver stuff for 2 whole days now (!), and from what I can tell, it's not possible to use Fedora as a vserver host? [00:44] erhm... meant 'bash-correct'... [00:44] That question brings Cisco (actually, just me) here [00:44] mygrant: hmm, why do you think so? [00:44] mygrant: I have a fedora vserver host [00:45] then I'm incorrect [00:45] need a modified kernel and the tools, of course [00:45] I've just been doing searches, and read that the nptl stuff in Fedora made vserver impossible [00:45] you could run Fedora in a vserver (like a Redhat 9 host), but not the other way around [00:46] one of fedoras aims is to be vanilla kernel compatible ... [00:46] MrBawb, are you patching the Fedora kernel, or the vanilla? [00:46] mygrant: vanilla [00:46] Guess I'll just have to work harder then [00:46] so fedora should, by definition, work with vanilla kernels ... and vice versa, everything else is a bug ;) [00:47] $ uname -r ; cat /etc/redhat-release [00:47] 2.4.24-vs1.22 [00:47] Fedora Core release 1 (Yarrow) [00:47] nice [00:47] That settles it! ;-) [00:48] MrBawb, you're not using the latest stable vs patch? [00:48] Doener (~doener@pD9E122D6.dip.t-dialin.net) left irc: Ping timeout: 501 seconds [00:48] mygrant: I've been lazy on this box :) [00:48] Any indications that that would cause problems, because that's what I tried [00:49] I wouldn't expect so [00:49] nope 1.24 isn't very different from 1.22 [00:49] I also have an FC box (actually, this one) running the development stuff (2.6 kernel, etc)... I understand that there are experimental vs patches for the 2.6 kernel [00:49] about 6 hunk, 2 lines changed each ... [00:50] We're thinking about using vserver to run multiple testing environments on less physical boxes here [00:51] yeah, same here :) [00:51] Any word on if it works with 64 bit stuff? [00:51] I think someone was working on it [00:51] what 64 bit stuff? [00:51] That would also be useful [00:52] 64 bit kernels with 64 bit processors [00:52] like RHEL64 [00:52] sure, running on parisc64 for example (hppa64) [00:52] or SUSE 64 [00:52] well, if you throw out the suse kernel, sure ... why not? [00:52] no, x86 stuff [00:53] ia64 or x86_64 [00:53] x86_64 [00:53] is tested, should work, although IIRC last test was on 2.4.23 and vs1.20 [00:53] I think we'd be willing to be guinea pigs [00:54] We'd like to test out multiple Redhat+Oracle environments, but we only have one 64-bit box for now [00:54] mygrant: ia64? [00:55] Ahh, x86_64... I missed that. [00:55] Zoiah, yep, x86_64 [00:56] I need to get myself any of those boys also. Colleage of mine played with Oracle 9i on IA-64 and wasn't too impressed with it. [00:56] well, thing is, you can't test a RH kernel with vserver, but you can test a RH distro with it, for the kernel test case, you should use UML, QEMU or VMWARE ... [00:56] Bertl, oh yeah, I forgot that [00:57] We might have to use UML in that case [00:57] Though I don't like the performance hit [00:57] mygrant: I would guess the redhat people wont enjoy you patching their enterprise tested kernels with more stuff. [00:57] mygrant: vmware would keep the kernel more clean. [00:57] We wouldn't use it in production, of course :-) [00:58] Ofcourse, but still, I'm not sure if UML would apply to such a heavily patched kernel. [00:58] But it looks like that's out anyway [00:58] I guess we'll just have to order more boxes [00:58] ;-) [00:59] mygrant: #uml is also on this irc :) [00:59] well, we do not even know what you want to test, maybe other solutions would be more appropriate ... [00:59] Useful, this IRC [01:00] Bertl, I still want to play with Vserver for other uses [01:00] It just got me thinking [01:00] mygrant: it has far more uses than you would expect at first. :) [01:01] so patching the vanilla sources with the vserver stuff from linux-vserver.org should work on an FC1 box... [01:01] (I'm back to the beginning now) [01:04] it should [01:04] well, I'll be sure to report back [01:05] Any good how-to guides lying around? [01:05] I'm looking at this one currently: http://www.linux-vserver.org/index.php?page=New-Howto [01:05] Looks pretty good... [01:06] MrBawb, did you apply any other patches besides the straightup vserver one? [01:06] mygrant: none [01:07] http://mirror.datapipe.net/local/fedora-core-1/i386/ [01:07] I have the util-vserver rpms there if you'd like [01:07] excellent... thanks [01:10] Topic changed on #vserver by Bertl!~herbert@MAIL.13thfloor.at: http://linux-vserver.org/ || latest stable 1.24, devel 1.3.6, exp 0.04 [01:11] I shall return with good news! [01:11] mugwump (~sv@202-0-63-18.adsl.paradise.net.nz) joined #vserver. [01:11] hi sam! [01:11] mygrant (~mygrant@migo.cisco.com) left irc: Quit: Vserver! [01:12] Hi Herb. *sigh* another year, another failed venture :( [01:12] hmm ... is that so? [01:13] yeah, got kicked out over a contract dispute [01:13] the contract says that I can telework... [01:13] they signed it... [01:13] and now I've been given 7 days to make it all the way around the world. [01:14] Well, at least I'll now have CFT [01:14] CFT means? [01:17] Copious Free Time [01:18] ah, well, so you are now working full time vserver, right? [01:19] Heh. Well, I want to tidy up a few projects first, but there will be tmie for vserver no doubt [01:19] How's the 2.6 port going? [01:19] slowly ... [01:20] not too much demand, so I do not put too much time into it ... [01:20] fair enough [01:20] and I'm trying to fix things up in 2.4 before I port them ;) [01:20] Probably a good plan [01:24] Dammit ... need a new cvs repos :) [01:33] noel- (~noel@pD9E09913.dip.t-dialin.net) joined #vserver. [01:37] noel (~noel@p50859979.dip.t-dialin.net) left irc: Ping timeout: 504 seconds [02:25] WSU (~Josh@ny.webpipe.net) left irc: Quit: Leaving [03:04] Action: mids got vserver working! [03:05] great! [03:05] it wasnt that hard [03:06] though the process walkthrough of UML was a bit more friendly to me [03:07] hmm, please elaborate ... [03:07] hm [03:07] I had to get info from a lot of seperated docs for vserver [03:07] each telling little bits [03:07] ah okay, so the UML docu is better, right? [03:08] a bit more coherent [03:08] otoh, vserver has a wiki [03:08] I can understand that, we had no time to write documentation yet, it's all what people contributed ;) [03:08] so I could add all the missing features :) [03:08] it would be appreciated ... [03:10] the debian-newvserver.sh is great [03:10] this is with debottstrap, right? [03:10] yeah [03:10] s/bot/boot/ [03:11] actually s/bot/boo/ ;) [03:11] yes, I did that some time ago, and it is very simple, and useful ... [03:12] I have to figure out how to do the quotas [03:12] it seems that I need some additional kernel patch [03:12] well, depends on what you want to get ... [03:12] but those that I found were for =< 2.4.22 [03:12] hm, I want: [03:12] http://www.13thfloor.at/vserver/project/ [03:13] http://www.13thfloor.at/vserver/s_addons/quota/ [03:13] yeah, please continue ... [03:13] (also works for 1.24?) [03:14] - limit the disk usage for the virtual root of the whole vserver [03:14] should, havent tried, but if it fails, let me know, I'll update it ... [03:14] - have quota support within the vserver for individual users [03:14] well, in this case, you have to go for them ... ;) [03:15] kool, so it is both possible [03:15] basically your options are: [03:15] a) put each vserver on a separate partition (lvm, loop, real) [03:15] ah, doh... loop would be cool for that [03:15] in this case, you do not need any additional patches, but loose the unification and caching ... [03:15] (why didnt i think of it, that is exactly how it works with UML) [03:16] b) use the q0.12 patch, and have live adjustable disk limits and user/group quotas inside the vservers [03:17] this works on shared partitions (one partition for several vservers) [03:17] so you gain the unification and inode cache bonus ... [03:17] nice [03:18] only mandatory thing is, you have to use static context ids ... [03:18] and I would strongly advise, that you put the vservers on a separate partition, not the root partition ;) [03:18] I have another HD somewhere [03:18] that would work [03:19] would it also be possible to limit memory usage of the vserver somehow? [03:19] probably though ulimit [03:20] http://www.13thfloor.at/vserver/d_release/overview [03:20] but so far all ulimit tweaks gave errors (I bet due to some restriction locs) [03:20] " memory accounting and resource limits" [03:20] there we go [03:20] well, I guess your config uses ULIMIT="-H -u 100" [03:20] where it should use ULIMIT="-HS -u 100" [03:21] yeah [03:21] this has changed in the vanilla kernel around 2.4.23 .. and most configs didn't check it ;) [03:21] last question: do you also accept e-gold donations besides paypal? [03:22] what is e-gold? [03:22] (paypal is just too evil) [03:22] e-gold is a digital gold currency [03:22] http://www.e-gold.com/ [03:22] hmm, let me have a look ... [03:22] mozilla, freenet and some other opensource projects take it [03:23] (paypal is very bad on the privacy, high on fees, have crooks limiting accounts at random, etc) [03:23] I know, I know ... [03:23] ((it works for most ppl, until it goes wrong, see paypalsucks.com etc) [03:23] ) [03:23] let me check that one ... [03:24] no hurry, my gold wont run away [03:24] nor me :) [03:28] hmm, what can you do with e-gold? can you withdraw it, or transfer the value to a bank account? [03:30] not directly (unless you have like 400oz of gold), so you normally sell it through an exchange provider (there are a lot of them, some doing wires, cheques, cash, etc) [03:30] hmm, does this work in europe too? [03:30] I live in the Netherlands [03:31] normally I use icegold (who are kickass), they are based in estonia, there are some french / german ones too [03:33] problem is that using creditcards is a lot easier [03:33] (but you have fraud problems, insecurity, bla) [03:36] anyway, take a look, if it might be something for you, let me know and I'll donate some gold; otherwise no hard feelings; I'll try to compensate with wiki edits (once I understand it a bit) :) [03:36] I'm trying to get an account ;) [03:38] 'Cheap, Fast, Good choose any two.' hehe [03:38] Account 1193801 [03:39] but helping with the wiki is also a good option ;) [03:40] ok, you got EUR 25 [03:41] okay and 25 cent are transfer fee, I guess ... [03:41] now dont forget your password :) [03:42] (fees, yeah) [03:42] thanks for your donation ... [03:43] Bertl: as you mention the wiki, what's the idea behind the change from https to http? i had little time to have a closer look, so asking will be much easier ;) [03:43] maybe more next time ;) I just got a bugbounty of $100 offered. [03:43] night [03:43] have a good one! [03:44] Doener``: hmm, what https/http change? [03:44] ah, you mean in my modifications? [03:44] iirc it was a url pattern [03:44] exactly :) [03:45] yeah, that has no reason at all, was just a change I did to test something (the pattern matching) and forgot to change back ;) [03:45] ah, i thought there was more behind it as you mentioned it not to be the 'final solution' [03:47] well, thing is, we had interwiki links enabled, and the matching done in tavi, resulted in each time stamp 12:13:14 becoming an interwiki link ;) [03:48] I made I cruel hack attempt, to fix this in a tavi appropriate manner, but failed, as the pattern is wildly mixed with other patterns all over the code ... [03:54] hm... although i don't know why, that just caught my interest... will drop my current work and work on tavi ;) [04:22] interesting to see e-gold mentioned. From experience, Pecunix are better. [04:22] Any e-currency vendor that lets you confirm transactions by PGP signing text certainly get my vote ;-) [06:08] ccooke (~ccooke@spc1-walt1-4-0-cust238.lond.broadband.ntl.com) joined #vserver. [06:08] hi ccooke! [06:09] re [06:09] hello [06:10] hmm, kestrel, sneaked in ;) [06:11] Action: kestrel is very sneaky [06:11] yeah, I guess I underestimated your sneakiness 8-) [06:12] hehehe :) [06:15] guess you guys are here, because you want something to test for the weekend, right? [06:24] okay, guess I'm wrong ;) [06:33] do you have a new release coming herbert? [06:34] well, there is 1.3.6 out there, plenty of new features, and there will be a 2.6.2-rc1 release verry soon ;) [06:45] ooh [06:45] i'm still running 1.3.4 [06:45] Action: kestrel should upgrade [06:47] ccooke (~ccooke@spc1-walt1-4-0-cust238.lond.broadband.ntl.com) left irc: Ping timeout: 501 seconds [06:56] hahaha coke wasread cook [06:56] of, beer, wine and me and booze = mix [06:56] hmm, hi click! [06:59] heh [07:00] drunk click [07:07] hela [07:08] yup, durnk, not very, but ok [07:08] uhm, too muhc [07:08] booze [07:09] i am good guy, but nooooo girl in mu bed [07:09] hum [07:09] a good [07:09] i miss my eex [07:09] good fu.. [07:09] and fun [08:26] mugwump_ (~sv@202-0-63-165.adsl.paradise.net.nz) joined #vserver. [08:31] mugwump (~sv@202-0-63-18.adsl.paradise.net.nz) left irc: Ping timeout: 492 seconds [08:32] kestrel (athomas@dialup51.optus.net.au) left irc: Ping timeout: 492 seconds [09:25] kestrel (athomas@dialup51.optus.net.au) joined #vserver. [09:33] hi there [09:39] hi! [09:39] (or not so high ;) [10:01] okay, have a good wossname, cu all later ... [10:03] Nick change: Bertl -> Bertl_zZ [10:22] ensc (~ircensc@ultra.csn.tu-chemnitz.de) left irc: Ping timeout: 480 seconds [11:07] mugwump_ (~sv@202-0-63-165.adsl.paradise.net.nz) left irc: Quit: leaving [11:07] serving- (~serving@213.186.190.24) left irc: Read error: Connection reset by peer [13:06] serving (~serving@213.186.190.24) joined #vserver. [13:18] Nick change: noel- -> noel [14:12] hm, the webserver is behaving weird [14:13] www.linux-vserver.org doesnt work; linux-vserver.org does [14:14] for me, both work [14:18] guess my nameserver is borked [14:20] guess so... [14:20] www.linux-vserver.org. 3595 IN CNAME linux-vserver.org. [14:20] linux-vserver.org. 3934 IN A 207.253.4.250 [14:21] they should never differ [14:25] restarting bind did the trick [15:16] loger joined #vserver. [15:31] miller7 (none@213.239.180.106) joined #vserver. [15:32] hello ppl [18:27] Nick change: Bertl_zZ -> Bertl [18:28] hi folks! [18:28] hey bert [18:37] who wants to do some linux 2.6.x testing? [19:13] ahh I see, guess I can put it on hold for another year then ;) [19:13] :) [19:16] hmm, greece ... [19:17] what does irismedia do? besides flash programming? [20:28] vat (vat@pD9E372BB.dip0.t-ipconnect.de) joined #vserver. [20:32] hi vat! [20:32] Topic changed on #vserver by Bertl!~herbert@MAIL.13thfloor.at: http://linux-vserver.org/ || latest stable 1.24, devel 1.3.6, exp 0.05 [20:34] arghs. [20:34] loadavg of 700. [20:34] because of lame apache. [20:35] 700 is not too shabby! [20:36] hmm. [20:36] our customers are very strange. [20:37] root 4725 37 vl08s15 11.2 1.1 12256 11396 /vservers/vl08s15/dev/pts/27 S 18:36 0:00 /usr/bin/dpkg --configure libdps1 libxaw7 xlibmesa3 xbase-clients [20:50] ? [21:00] installing X in a vserver.. [21:00] naaaaaaah. [21:00] idiots :) [21:01] not installing x tho [21:01] it's part of some other package [21:01] most probably pdf-support [21:01] which requires some xlibs etc [21:02] hi click, sober again? [21:02] yup [21:02] ready for a better dinner [21:02] click, there was also x-desktop-* ;) [21:02] and xserver and blah. [21:02] well, everything you need on a good server ... [21:02] what about hardware monitoring? ;) [21:03] configure: WARNING: ext2fs/ext2fs.h: present but cannot be compiled [21:04] wtf, my os is screwed :( [21:04] Report this to enrico.scholz@informatik.tu-chemnitz.de [21:04] sigh [21:05] what version of autoconf is needed for utils? [21:06] that is a broken ext2fs.h header ... [21:06] meaning? [21:07] best is to get a recent ext2fs.h ... there is a bugreport filed by enrico on that [21:07] https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=112448 [21:08] the additional comment should help you ;) [21:08] thank you [21:09] Bertl, I slept after 8 jim beam cola... [21:09] problem was a ddos on apache server :/ [21:13] may sound a stupid question, but how do I update ext2fs.h to latest version? where does it belong? [21:14] best would be to 'just' modify them by hand for now ... [21:14] but I can't find the lines enrico talks about! [21:16] okay, what 'packages' related to ext2 or e2fs do you ahve? [21:17] I wish I knew :-) [21:17] this gentoo thingie is too new for me [21:17] well, I use mandrake, the command there would be rpm -qa ext2 [21:18] for example, in my ext2fs.h file I have: [21:18] extern errcode_t ext2fs_resize_mem(unsigned long old_size, [21:18] unsigned long size, void **ptr); [21:18] EXT2FS_ATTR is not there [21:23] hmm, that looks good ... [21:24] but is this the file in ext2fs/ext2fs.h? [21:24] hmmm [21:25] _ /usr/include/ext2fs/ext2fs.h [21:25] there are two version, usually, one provided by the kernel, and another one provided by the libe2fs-devel ... [21:25] I don't think I have the devel one installed [21:26] not sure though cause for some reason my system's paths are gone bad [21:26] hmm, guess that is required, don't ask me what the package name is, though ... [21:26] don't worry, I'll find it [21:27] I think the RH EOL story will either make a lot of good in Linux or a lot of bad [21:27] cause people used to have all these things done for them [21:27] so now (like me) have to be used to more "inside linux" [21:27] I also have this: /root/vservers/tools/util-vserver-0.27/src/ext2fs.h [21:29] let me have a look at the config.log ... please .. [21:29] I will upload the new one as soon as my gentoo stops installing the upgrades [21:29] ok? [21:29] it still has 7 packages to compile [21:30] miller7: the RHL -> { RHEL, Fedora } split seems to have worked pretty ok [21:30] there are a few rough edges left, but the majority of the users seems to be fairly happy with it [21:30] RHEL needs $300 per CPU [21:30] Fedora: I haven't seen yet [21:31] white box linux may be interesting (RHEL with another name) [21:31] $300 per CPU ? where did you get that number ? [21:31] dunno [21:31] per machine or something [21:32] it all depends [21:33] the pricing seems to go from $180 to $18k per system without any discounts, depending on services platform etc [21:33] medium to large sized customers don't get anything without discounts ;) [21:33] I think HPC customers get like 90% discount [21:33] server farms should get a decent discount too, depending on the number of systems [21:33] hmm [21:35] though of course the sales folks make more money if they sell with less discounts, so you'd really need to negotiate everything ;) [21:35] ojn (~olof@cs6625217-161.austin.rr.com) joined #vserver. [21:35] hi ojn ! [21:35] hi [21:37] I have a question, maybe it's a faq but I didn't find anything about it... I have a machine with two interfaces, one internal and one external. If I create a vserver with addresses assigned on both interfaces, the second interface gets the right address and netmask but it gets the same broadcast as the first interface. [21:37] Obviously this causes problems with connectivity. :) [21:37] Is this a known problem / configuration mistake? [21:37] how does your config look like? [21:38] S_HOSTNAME="mail" [21:38] IPROOT="eth1:66.141.50.11/29 eth0:172.16.1.5/24" [21:38] ONBOOT="yes" [21:38] S_NICE="" [21:38] S_FLAGS="lock nproc" [21:38] ULIMIT="-H -u 256 -n 1024" [21:38] S_CAPS="CAP_NET_RAW" [21:38] what tool version? [21:39] good question. Hold on a sec. :) [21:40] http://vserver.13thfloor.at/Stuff/testme.sh [21:42] all but 201 passed [21:43] yeah, what version ;) [21:43] Linux 2.6.1 i686/chcontext 0.24/chbind 0.24 [J] [21:44] this is Jack's vserver-0.24 for example ;) [21:44] Linux 2.4.24-vs1.24 i686/chcontext 0.29/chbind 0.29 [J] [21:44] Doh. Still tired, I completely missed the version line :) [21:45] okay, this is vserver-0.29 ... [21:45] simplest way would be to setup the aliases yourself [21:46] ifconfig eth0:XXX 172.16.1.5 netmask 255.255.255.0 broadcast 172.16.1.255 [21:46] and if config eth1:XXX 66.141.50.11 netmask 255.255.255.248 [21:46] Yeah. But how do I keep the vserver scripts from adding a new alias on their own? [21:46] and if config eth1:XXX 66.141.50.11 netmask 255.255.255.248 broadcast 66.141.50.15 [21:47] and then use IPROOT="66.141.50.11 172.16.1.5" [21:47] leaving out the eth0/1 will not create the aliases ... [21:47] ok, thanks [21:47] http://vserver.13thfloor.at/Stuff/VServer-IP-Setup-0.1.txt [21:48] (see that for more details) [21:54] excellent, that did it. Thanks. [22:27] np [22:37] vat (vat@pD9E372BB.dip0.t-ipconnect.de) left irc: Ping timeout: 480 seconds [22:44] mhepp (~mhepp@r72s22p13.home.nbox.cz) joined #vserver. [23:08] what gcc version do you people use for Enrico's utils? [23:15] all dead :) [23:25] arekm (misiek@ikar.t17.ds.pwr.wroc.pl) joined #vserver. [23:34] hi all, did anyone tried patch for 2.6? i'm curious if there was any success with it [23:34] bert was looking for testers earlier [23:38] miller7: is main development focused on 2.6 or still 2.4? [23:38] 2.4 i think. not sure [23:40] uhm, my 2.4 kernel has 124 patches but I'll try to merge vserver, too and test it 8-) [23:40] sounds good ;) [23:40] hwhat gcc version / distro you have? [23:41] gcc 3.3.2, PLD Linux [23:41] u use enrico's utils? [23:41] well arekm, the 2.6.2-rc1 patch 0.05 should be fine ... [23:41] Bertl: ok, so I'll merge it with out distro 2.6 kernel and give it to users to they will test it 8-) [23:42] miller7: I'm not yet using vserver but willing to try [23:43] ah ok [23:43] some time ago I saw that Linus assigned syscal number for vserver - does that mean that vserver people will try to merge with mainline in 2.7 series? [23:44] well, maybe ... [23:44] I'm not sure that we want to 'merge' [23:45] interesting... [23:45] why not merging? [23:46] vserver is platform agnostic ... [23:46] but, we didn't manage to allocate a syscall for each platform for example ... [23:47] I know that a lot people just say - hey, the patch is here, you can use it, no need to merge but try to merge 124 patches with kernel, some very intrusive, that's hell 8-) [23:47] I know that, but even one patch, which doesn't merge, is often all you need ;) [23:49] by the way, what do you add with 124 patches? I assume most of them are small fixes, right? [23:50] mhepp (~mhepp@r72s22p13.home.nbox.cz) left irc: Remote host closed the connection [23:51] ah okay, you are merging 'the' PLD kernel ... i see [23:52] some are small, some not - xfs, acl, libata, O(1), grsecurity, ntfs, squashfs, ipp2p, imq, mppe, netfilter, device-mapper, new i2c, new aic7xxx, new acpi, new jfs [23:52] end more [23:52] yeah, most of the 'useful' patches are in Con's set at the moment ... [23:53] and vserver on ck1 is working, but I guess you don't want to include vserver 'unconditional' [23:53] http://cvs.pld-linux.org/cgi-bin/cvsweb/SPECS/kernel.spec?rev=1.581 [23:54] I want unconditional if it doesn't break anything when not used [23:54] hmm, well, it's intrusive, but it doesn't break the host system ... [23:55] at least it's not supposed to do, if it does, it's a bug ;) then please report it ;) [23:57] great [23:58] but I would suggest having a look at the development release, it's quite stable and somewhat advanced ... [23:59] oh, 2.6.1 patch doesn't apply to current bk. ,,they'' changed few lines in unistd.h doh. rediffing is booring [00:00] --- Sun Jan 25 2004