T2 IRC Log: 2007-10-08
This is the log as captured by an IRC bot in the channel. The statements are those of the individual people and might not neccessarily reflect the policy and legal rules as set forth by the T2 SDE Project.
--- Log opened Mon Oct 08 00:00:42 2007
00:17 -!- Enqlave|laptop [n=stealth@unaffiliated/enqlave] has quit ["There is intelligent life out there..."]
00:46 -!- Enqlave|laptop [n=stealth@unaffiliated/enqlave] has joined #t2
00:54 -!- sepp_ [n=sepp@Z7327.z.pppool.de] has joined #t2
01:14 -!- sepp [n=sepp@Z4fce.z.pppool.de] has quit [Read error: 110 (Connection timed out)]
01:37 -!- pdinoto_ [n=pdinoto@200.117.246.103] has quit [Read error: 104 (Connection reset by peer)]
01:38 -!- pdinoto_ [n=pdinoto@200.117.246.103] has joined #t2
02:31 -!- mqueiros_ [n=mqueiros@80.243.90.138] has quit []
03:15 -!- Key_Gena95 [i=lok@gateway/tor/x-b2ee396903283484] has joined #t2
04:36 -!- mqueiros [n=mqueiros@80.243.90.138] has joined #t2
04:54 -!- cgetty [n=elmo@pool-71-108-239-203.lsanca.dsl-w.verizon.net] has quit [Remote closed the connection]
07:09 -!- cgetty [n=clark@pool-71-108-239-203.lsanca.dsl-w.verizon.net] has joined #t2
07:16 -!- Enqlave|laptop [n=stealth@unaffiliated/enqlave] has quit ["There is intelligent life out there..."]
08:04 -!- cgetty [n=clark@pool-71-108-239-203.lsanca.dsl-w.verizon.net] has quit [Remote closed the connection]
09:18 < rxr> re
09:18 [Users #t2]
09:18 [@ChanServ] [ Dallur ] [ Fatal ] [ LMJ ] [ pdinoto_] [ sepp_ ]
09:18 [ CIA-8 ] [ emte ] [ happycube ] [ mqueiros] [ Ragnarin] [ TobiX ]
09:18 [ cyda ] [ Enqlave] [ Key_Gena95] [ mtr ] [ rxr ] [ valentin]
09:18 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
09:32 < rxr> Ragnarin: awake ?
09:34 < CIA-8> rene * r26256 /trunk/ (3 files in 3 dirs):
09:34 < CIA-8> * fixed sg (was ag) typo in 2 early userspace init scripts
09:34 < CIA-8> * unified them more (no hardcode module block, use a for loop)
09:34 < CIA-8> * updated year and improved status printout to include an indication
09:34 < CIA-8> which of the init variants is running
10:16 -!- Enqlave|laptop [n=stealth@unaffiliated/enqlave] has joined #t2
10:26 -!- Enqlave|laptop is now known as Enqlave|xps
10:28 -!- Enqlave|xps is now known as [Enqlave]
10:31 < CIA-8> rene * r26257 /trunk/package/network/bluez-libs/bluez-libs.desc: * updated bluez-libs (3.19 -> 3.20)
10:31 < CIA-8> rene * r26258 /trunk/package/network/bluez-utils/bluez-utils.desc: * updated bluez-utils (3.19 -> 3.20)
10:32 < CIA-8> rene * r26259 /trunk/package/textproc/iso-codes/iso-codes.desc: * updated iso-codes (1.4 -> 1.5)
10:32 < CIA-8> rene * r26260 /trunk/package/perl/perl-yaml/perl-yaml.desc: * updated perl-yaml (0.65 -> 0.66)
10:32 < CIA-8> rene * r26261 /trunk/package/develop/python-xlib/python-xlib.desc: * updated python-xlib (0.13 -> 0.14)
10:32 < CIA-8> rene * r26262 /trunk/package/python/pyrexc/pyrexc.desc: * updated pyrexc (0.9.5.1a -> 0.9.6)
10:37 < CIA-8> rene * r26263 /trunk/package/e16/enlightenment16/enlightenment16.desc: * updated enlightenment16 (0.16.8.9 -> 0.16.8.10)
10:48 -!- Key_Gena95 [i=lok@gateway/tor/x-b2ee396903283484] has quit [Excess Flood]
11:07 -!- [Enqlave] is now known as Enq|xps
11:08 [Users #t2]
11:08 [@ChanServ] [ Dallur ] [ Enq|xps ] [ LMJ ] [ pdinoto_] [ sepp_ ]
11:08 [ CIA-8 ] [ emte ] [ Fatal ] [ mqueiros] [ Ragnarin] [ TobiX ]
11:08 [ cyda ] [ Enqlave] [ happycube] [ mtr ] [ rxr ] [ valentin]
11:08 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
11:41 < rxr> ok, increasins read-ahead helped LVM2+RAID5 performance a little
12:01 -!- Enq|xps [n=stealth@unaffiliated/enqlave] has quit ["There is intelligent life out there..."]
12:18 < rxr> Ragnarin: do you know why our util-vserver package is named utils-vserver (not the plural of util)
12:18 < rxr> Ragnarin: upstream it is just util-vserver ...
12:47 -!- TobiX [i=tobias@zoidberg.org] has quit [Remote closed the connection]
13:44 -!- mqueiros [n=mqueiros@80.243.90.138] has quit []
14:19 < CIA-8> rene * r26264 /trunk/architecture/share/kernel-fs.conf.m4: * changed shared kernel configuration to enable ACL/XATTR by default
14:32 < CIA-8> rene * r26265 /trunk/package/develop/autogen/autogen.desc: * updated autogen (5.9.2 -> 5.9.3)
14:32 < CIA-8> rene * r26266 /trunk/package/textproc/msort/msort.desc: * updated msort (8.40 -> 8.42)
15:27 < CIA-8> mtr * r26267 /trunk/package/base/uclibc/ (internal_function.patch patches/gettext-redefinition.patch):
15:27 < CIA-8> * added uClibc hotfix for internal_function declaration, this is
15:27 < CIA-8> internal and has not be exposed to user applications
15:27 < CIA-8> (uClibc svn changeset 18907)
16:02 < rxr> hm - damn vserver does not support mknod in the containers, this sucks
16:03 < rxr> and disqualifies them for building T2 within ... :-(
16:29 < CIA-8> rene * r26268 /trunk/package/www/owb/ (. compile.patch owb.cache owb.conf owb.desc): * added owb (1.0) - Origyn Web Browser designed for CE devices
16:34 -!- Dallur [n=earl@fgfw.basis.is] has quit [Read error: 110 (Connection timed out)]
16:37 -!- Key_Gena95 [i=lok@gateway/tor/x-fe2e366434bb8660] has joined #t2
16:48 < CIA-8> rene * r26269 /trunk/package/games/libkxl/libkxl.desc: * removed duplicate CROSS tak from libkxl
16:51 < CIA-8> rene * r26270 /trunk/package/emulators/hercules/hercules.desc: * removed duplicate CROSS tag from hercules, likewise
17:12 [Users #t2]
17:12 [@ChanServ] [ emte ] [ happycube ] [ mtr ] [ rxr ]
17:12 [ CIA-8 ] [ Enqlave] [ Key_Gena95] [ pdinoto_] [ sepp_ ]
17:12 [ cyda ] [ Fatal ] [ LMJ ] [ Ragnarin] [ valentin]
17:12 -!- Irssi: #t2: Total of 15 nicks [1 ops, 0 halfops, 0 voices, 14 normal]
17:20 < pdinoto_> hi rene!
17:20 < pdinoto_> just wanted to let you know that I am using my T2 "micro" package, which is based
17:21 < pdinoto_> on tresc. It is a tiny X system with networking and lua... boots in 5 secs
17:21 < pdinoto_> on a pentium board, runs on 18mb of ram (x with 4 xterms opened)
17:21 < rxr> nice :-)
17:21 < pdinoto_> and is under 32mb of disk uncompressed. Yay!
17:22 < pdinoto_> s/package/target/
17:22 < rxr> can you publish it so others can enjoy it, too ?
17:23 < pdinoto_> sure... it is a little bit messy (I keep playing with package list as I need them).
17:23 < pdinoto_> Do I send the files to your email?
17:25 < rxr> yes, or the t2 list: t2@t2-project.org
17:51 < CIA-8> rene * r26271 /trunk/package/graphic/tesseract/tesseract.desc: * updated tesseract (2.00 -> 2.01), added CV-URL tag
18:00 < rxr> Ragnarin: btw. the ide-generic is already loaded last, hm ?
18:00 < CIA-8> rene * r26272 /trunk/package/base/mkinitrd/mkinitrd.sh: * include chroot in the default initrd, just handy to be around
18:02 < CIA-8> mtr * r26273 /trunk/package/ (3 files in 2 dirs): * updated x264 (20061104-2245 -> 20071007-2245)
18:03 < CIA-8> mtr * r26274 /trunk/package/multimedia/ffmpeg/ (ffmpeg.desc libpostproc-destdir.patch): * updated ffmpeg (2007-04-25 -> 2007-10-07)
18:44 -!- Dallur [n=earl@fgfw.basis.is] has joined #t2
18:45 -!- Dallur [n=earl@fgfw.basis.is] has quit [Remote closed the connection]
19:34 -!- cgetty [n=Clark@pool-71-108-239-203.lsanca.dsl-w.verizon.net] has joined #t2
19:36 < Ragnarin> rxr: oh it is? hmmmm
19:37 < rxr> Ragnarin: I posted quite some stuff to you today ...
19:37 < rxr> did you got all scrollback ?
19:38 < Ragnarin> yea
19:38 < rxr> first of all did you notice the util-vserver is named utils-vserver in T2 ?
19:38 < rxr> is this useful for anything but global confusion =?
19:38 < rxr> can we fix this ?
19:38 < Ragnarin> more confusing is that we have vserver-utils and utils-vserver
19:39 < Ragnarin> we should
19:39 < rxr> yeah - but this vserver-utils is this new thing ...
19:39 < rxr> I think that naming is ok, as it is upstream ...
19:39 < rxr> do you use vserver ?
19:40 < Ragnarin> I try to ... trying to get a test system running
19:40 < rxr> what do you wanna use it for ?
19:41 < Ragnarin> user isolation
19:41 < rxr> ah
19:41 < rxr> I got a guest up today, works so far, _but_ vserver does not virtualize /dev/*
19:41 < Ragnarin> you can do that with grsec tho
19:41 < rxr> so it forbids mknod() so the guest containers can not create new ones
19:42 < rxr> oh
19:42 < Ragnarin> the guests shouldn't be able to create new dev nodes anyway?
19:42 < rxr> ok, I wanted to get openvz packaged for T2, now because it looks like opnevz also allows restricting device node access (and not just forbid mknod)
19:43 < rxr> actually I only need this lightweight containers to give balzius a partition of our new 8-core monser ...
19:43 < rxr> for t2/security work ...
19:43 < Ragnarin> ah ok :)
19:43 < rxr> so I need somehting where mknod and chroot work ...
19:43 < Ragnarin> chroot should work
19:43 < rxr> I package openvz now, we should have it anyway
19:43 < rxr> let's now how this turns out ...
19:44 < rxr> who renames this utils-verver? do you want to ?
19:44 < Ragnarin> do you know how? :)
19:44 < rxr> rename ?
19:45 < Ragnarin> can I do that on the whole dir?
19:45 < Ragnarin> without messing up the history?
19:45 < rxr> ah yes, files inside a dir and a dir in one svn commit was tricky
19:45 < rxr> I'll figure it out and post the comments "how to" to our svn page
19:46 < Ragnarin> ok good :)
19:47 < rxr> regarding the ide-generic, it is after the udevstart / udevtrigger thing
19:47 < rxr> it should be loaded after the udev stuff
19:47 < Ragnarin> btw does dev.croup.de resolve for you?
19:47 < rxr> though - maybe - we should insert udevsettle as barrier so it is not loaded quicker than the udev even processing takes ...
19:47 < Ragnarin> are you sure the autodetection doesn't load it earlier?
19:48 < rxr> oh - you mean tha auto-detection loads generic-ide ?
19:48 < rxr> that could be, yes
19:48 < rxr> Unknown host dev.croup.de
19:48 < Ragnarin> I think so
19:49 -!- cgetty [n=Clark@pool-71-108-239-203.lsanca.dsl-w.verizon.net] has quit ["Leaving"]
19:50 < Ragnarin> hmm so vserver-utils is dead ... *googles*
19:51 -!- kic- [i=rfoote@208.115.42.120] has joined #t2
19:51 < kic-> re
19:51 < kic-> rxr leave already?
19:52 < rxr> nope
19:52 < Ragnarin> no
19:52 < rxr> hie
19:52 < kic-> wee
19:52 < kic-> heh
19:52 < kic-> hi
19:52 < rxr> still fighting virtual server crap
19:52 < kic-> ah
19:52 < Ragnarin> lol
19:52 < rxr> did you test the new image s?
19:52 < kic-> yanno those changes you made to the livecd/init that you emailed me about a little bit ago?
19:52 < kic-> not yet
19:53 < rxr> hm ?
19:53 < rxr> your second last post was too crypted for me .-)
19:54 < CIA-8> rene * r26275 /trunk/package/security/ (8 files in 2 dirs):
19:54 < CIA-8> * renamed utils-vserver -> util-vserver, as that is the upstream
19:54 < CIA-8> name and everything else is just confusing
19:54 < kic-> he changed his mind, and said he wouldn't mind waiting until you were able to get a good solution for it
19:54 < kic-> if it only took a day or two
19:54 < rxr> kic-: but the quickfix should be ready to test
19:55 < rxr> kic-: I could not test, as I can not copy the flash intended files over for quikc check, but I hope it works, and it is done now anyway ...
19:55 < rxr> it just needs a test run (and maybe a quick-fix interation in the case I made a typo :... )
19:56 < kic-> right..
19:57 < kic-> how difficult though is it to revert back?
19:57 < kic-> if the test does not go well?
19:57 < rxr> just a single file that is not even part of your target
19:57 < rxr> it turned out to be a two-liner
19:57 < kic-> ah
19:57 < kic-> nice
19:58 < kic-> did you back up old file? or hash out or something?
19:58 < rxr> but as it is not part of your target it's a little bad for versioning
19:58 < rxr> svn revert on the file will do :-)
19:59 < kic-> ugh
19:59 < kic-> hehe
19:59 < kic-> which file?
20:00 < rxr> target/share/livecd/init2
20:02 < rxr> ouh - damn
20:02 < rxr> something conflicts with the openvz patch, and I fear it's our t2 no2mod stuff
20:02 < rxr> damnit
20:02 < rxr> t
20:02 -!- mqueiros [n=mqueiros@80.243.90.138] has joined #t2
20:03 < rxr> ieeek
20:03 < rxr> they patch-in a don_ask mode ??!?!?!?!
20:03 < rxr> damn pseudo open source companies
20:03 < rxr> I know why I tried vserver before openvz ...
20:04 < rxr> I wonder if we need our no2mod patch at all - maybe it would just work if we yes 'm' | make oldconfig or so ..
20:09 < rxr> too bad atop conficts with openvz, likewise
20:13 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has joined #t2
20:16 -!- mqueiros [n=mqueiros@80.243.90.138] has quit []
20:16 < tri> moin
20:17 < rxr> moin tri
20:20 -!- Enq|xps [n=stealth@unaffiliated/enqlave] has joined #t2
20:26 < kic-> hrm, new jffs2 version out
20:27 < rxr> :-)
20:27 < rxr> you wanna patch it in ?
20:27 < kic-> would be wonderful if it will solve any of my headaches :D
20:28 < kic-> is that a kernel patch?
20:28 < tri> hi rene any news about the xen support in t2 ?
20:28 < tri> t2 as dom0 ?
20:28 < rxr> tri: with some manual work here and there yet
20:28 < tri> manuel work ?
20:28 < rxr> but as my interested in Xen dropped before I finished integration patches welcome ...
20:29 < tri> last time i wasn't able to build it :-)
20:29 < rxr> you have to build the kernel manually ...
20:29 < tri> why was your interested in xen dropped ?
20:29 < rxr> not dropped
20:29 < rxr> just not finished
20:29 < tri> :-)
20:30 < rxr> I'm on Linus and Alans side on this: an additional, crappy scheduler below linux just doesn't cut
20:30 < rxr> it
20:30 < rxr> runing the containers as Linux processes is more viable ..
20:30 < tri> i have the change to test xen on a big machine (quadcore 16gb ram) on wednesday
20:31 < tri> ah ok does kvm or something else work different there ?
20:31 < tri> or what do you mean as "containers" ?
20:32 < rxr> yes, kvm
20:32 < rxr> another thing you should consider: the fat xen machines have fixed memory patition as there are full OS instances in there
20:32 < rxr> tri: when you just have linux servers, using vserver or openvz saves you a lot of memory as the memory is dynamically allocated between all your virtual containers
20:33 < rxr> with something like xen / vmware you waste a lot of memory not only on the kernel's and their environment (library loaded etc.) but aos each VM will have free memory that is not shared among them
20:37 < tri> hm but with vserver i don't have a change to run win / netware and linux on one hardware
20:37 < tri> how about kvm
20:37 < tri> do you thing the idea behind kvm is better then the xen idea ?
20:38 < rxr> well - atleast for the "reusing the Linux process scheduler and memory management"
20:38 < rxr> it has less management tools around, though - however, I don't like the Xen python bloat ... so that might be a + :-)
20:38 < rxr> maybe check them all out
20:38 < tri> :-)
20:38 < rxr> and if you can finish Xen integration in T2 :-)
20:39 < tri> hm yeah i will try to use t2 - but i think i have to use novell linux - "standard company linux"
20:40 < tri> :-(
20:41 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has left #t2 []
20:41 -!- pdinoto_ [n=pdinoto@200.117.246.103] has quit [Read error: 104 (Connection reset by peer)]
20:41 -!- pdinoto_ [n=pdinoto@200.117.246.103] has joined #t2
20:46 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has joined #t2
20:48 < tri> re
20:48 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has left #t2 []
20:48 < rxr> damnit
20:48 < rxr> at least the vserver code compiled out of the box ...
20:48 * rxr slabs this rusian SWSoft el-cheapo coders ...
20:50 < rxr> 2.6.22 ovz kernel is in early alpha, and chechpointing functionality is
20:50 < rxr> not yet fully ported. For now, you have to deselect it if you want to
20:50 < rxr> try compiling that kernel.
20:50 < rxr> hm
20:54 -!- mqueiros_ [n=mqueiros@80.243.90.138] has joined #t2
21:10 < rxr> btw:
21:10 < rxr> http://people.exactcode.de/~rene/_G0D6747.jpg
21:10 < rxr> http://people.exactcode.de/~rene/_G0D6749.jpg
21:12 [Users #t2]
21:12 [@ChanServ] [ emte ] [ Fatal ] [ kic- ] [ mtr ] [ rxr ]
21:12 [ CIA-8 ] [ Enqlave] [ happycube ] [ LMJ ] [ pdinoto_] [ sepp_ ]
21:12 [ cyda ] [ Enq|xps] [ Key_Gena95] [ mqueiros_] [ Ragnarin] [ valentin]
21:12 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
21:17 < sepp_> i c red
21:28 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has joined #t2
21:43 < LMJ> Nice
21:45 < tri> re
21:46 < LMJ> NICE
21:49 < tri> what is nice ?
21:50 < mqueiros_> tri: I believe LMJ was talking about this: http://people.exactcode.de/~rene/_G0D6747.jpg and http://people.exactcode.de/~rene/_G0D6749.jpg
21:50 < LMJ> ;)
21:52 < tri> cool :-)
21:52 < tri> my girlfriend
21:55 < kic-> thats not your gf
21:55 < kic-> :D
21:56 < tri> ah ok, right :-)
21:56 < kic-> is it?
21:56 < kic-> lol
21:56 < tri> do you know that shirt "you need a girfriend without *.jpg in here name"
21:57 < kic-> no
21:57 < kic-> they make a shirt that says that?
21:57 < kic-> i want one
21:57 < kic-> heh
21:57 < tri> :-)
21:57 < kic-> omg, i know i'm looking at to much code when i browse some C code for jffs2 patch in the kernel, and i begin to understand what it means =(
21:58 < kic-> apparently god is going to force me to learn C one way or another
21:58 < tri> hm how many mem can i allocat at one time under linux ? (x86)
21:59 < kic-> *shrug* the full amount of your ram chip?
21:59 < kic-> heh
21:59 < kic-> can't you vmalloc the whole thing?
21:59 < kic-> if you wanted to :D
22:04 < Ragnarin> there's a ulimit setting usually
22:05 < tri> hm the ulimit setting on my box is unlimited
22:05 < tri> can i allocate as many i like to ?
22:05 < Ragnarin> then you'd be limited by the address space you get per process
22:06 < Ragnarin> which can be limiting on 32bit at least
22:06 < tri> what happens if i allocate "to much" - is this possible
22:07 < tri> i have read something about oom killer
22:07 < Ragnarin> well look at for example the ace package configure script (among others)
22:08 < Ragnarin> when the machine runs out of virtual memory (real memory + swap) some processes will be killed
22:10 < tri> hm http://kerneltrap.org/node/1010
22:28 -!- TobiX [i=tobias@zoidberg.org] has joined #t2
22:29 < Ragnarin> seems the problem there is that the kernel gives out memory that's not actually there ... that's bad
22:50 < tri> gn8
22:50 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has left #t2 []
--- Log closed Tue Oct 09 00:00:42 2007
00:17 -!- Enqlave|laptop [n=stealth@unaffiliated/enqlave] has quit ["There is intelligent life out there..."]
00:46 -!- Enqlave|laptop [n=stealth@unaffiliated/enqlave] has joined #t2
00:54 -!- sepp_ [n=sepp@Z7327.z.pppool.de] has joined #t2
01:14 -!- sepp [n=sepp@Z4fce.z.pppool.de] has quit [Read error: 110 (Connection timed out)]
01:37 -!- pdinoto_ [n=pdinoto@200.117.246.103] has quit [Read error: 104 (Connection reset by peer)]
01:38 -!- pdinoto_ [n=pdinoto@200.117.246.103] has joined #t2
02:31 -!- mqueiros_ [n=mqueiros@80.243.90.138] has quit []
03:15 -!- Key_Gena95 [i=lok@gateway/tor/x-b2ee396903283484] has joined #t2
04:36 -!- mqueiros [n=mqueiros@80.243.90.138] has joined #t2
04:54 -!- cgetty [n=elmo@pool-71-108-239-203.lsanca.dsl-w.verizon.net] has quit [Remote closed the connection]
07:09 -!- cgetty [n=clark@pool-71-108-239-203.lsanca.dsl-w.verizon.net] has joined #t2
07:16 -!- Enqlave|laptop [n=stealth@unaffiliated/enqlave] has quit ["There is intelligent life out there..."]
08:04 -!- cgetty [n=clark@pool-71-108-239-203.lsanca.dsl-w.verizon.net] has quit [Remote closed the connection]
09:18 < rxr> re
09:18 [Users #t2]
09:18 [@ChanServ] [ Dallur ] [ Fatal ] [ LMJ ] [ pdinoto_] [ sepp_ ]
09:18 [ CIA-8 ] [ emte ] [ happycube ] [ mqueiros] [ Ragnarin] [ TobiX ]
09:18 [ cyda ] [ Enqlave] [ Key_Gena95] [ mtr ] [ rxr ] [ valentin]
09:18 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
09:32 < rxr> Ragnarin: awake ?
09:34 < CIA-8> rene * r26256 /trunk/ (3 files in 3 dirs):
09:34 < CIA-8> * fixed sg (was ag) typo in 2 early userspace init scripts
09:34 < CIA-8> * unified them more (no hardcode module block, use a for loop)
09:34 < CIA-8> * updated year and improved status printout to include an indication
09:34 < CIA-8> which of the init variants is running
10:16 -!- Enqlave|laptop [n=stealth@unaffiliated/enqlave] has joined #t2
10:26 -!- Enqlave|laptop is now known as Enqlave|xps
10:28 -!- Enqlave|xps is now known as [Enqlave]
10:31 < CIA-8> rene * r26257 /trunk/package/network/bluez-libs/bluez-libs.desc: * updated bluez-libs (3.19 -> 3.20)
10:31 < CIA-8> rene * r26258 /trunk/package/network/bluez-utils/bluez-utils.desc: * updated bluez-utils (3.19 -> 3.20)
10:32 < CIA-8> rene * r26259 /trunk/package/textproc/iso-codes/iso-codes.desc: * updated iso-codes (1.4 -> 1.5)
10:32 < CIA-8> rene * r26260 /trunk/package/perl/perl-yaml/perl-yaml.desc: * updated perl-yaml (0.65 -> 0.66)
10:32 < CIA-8> rene * r26261 /trunk/package/develop/python-xlib/python-xlib.desc: * updated python-xlib (0.13 -> 0.14)
10:32 < CIA-8> rene * r26262 /trunk/package/python/pyrexc/pyrexc.desc: * updated pyrexc (0.9.5.1a -> 0.9.6)
10:37 < CIA-8> rene * r26263 /trunk/package/e16/enlightenment16/enlightenment16.desc: * updated enlightenment16 (0.16.8.9 -> 0.16.8.10)
10:48 -!- Key_Gena95 [i=lok@gateway/tor/x-b2ee396903283484] has quit [Excess Flood]
11:07 -!- [Enqlave] is now known as Enq|xps
11:08 [Users #t2]
11:08 [@ChanServ] [ Dallur ] [ Enq|xps ] [ LMJ ] [ pdinoto_] [ sepp_ ]
11:08 [ CIA-8 ] [ emte ] [ Fatal ] [ mqueiros] [ Ragnarin] [ TobiX ]
11:08 [ cyda ] [ Enqlave] [ happycube] [ mtr ] [ rxr ] [ valentin]
11:08 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
11:41 < rxr> ok, increasins read-ahead helped LVM2+RAID5 performance a little
12:01 -!- Enq|xps [n=stealth@unaffiliated/enqlave] has quit ["There is intelligent life out there..."]
12:18 < rxr> Ragnarin: do you know why our util-vserver package is named utils-vserver (not the plural of util)
12:18 < rxr> Ragnarin: upstream it is just util-vserver ...
12:47 -!- TobiX [i=tobias@zoidberg.org] has quit [Remote closed the connection]
13:44 -!- mqueiros [n=mqueiros@80.243.90.138] has quit []
14:19 < CIA-8> rene * r26264 /trunk/architecture/share/kernel-fs.conf.m4: * changed shared kernel configuration to enable ACL/XATTR by default
14:32 < CIA-8> rene * r26265 /trunk/package/develop/autogen/autogen.desc: * updated autogen (5.9.2 -> 5.9.3)
14:32 < CIA-8> rene * r26266 /trunk/package/textproc/msort/msort.desc: * updated msort (8.40 -> 8.42)
15:27 < CIA-8> mtr * r26267 /trunk/package/base/uclibc/ (internal_function.patch patches/gettext-redefinition.patch):
15:27 < CIA-8> * added uClibc hotfix for internal_function declaration, this is
15:27 < CIA-8> internal and has not be exposed to user applications
15:27 < CIA-8> (uClibc svn changeset 18907)
16:02 < rxr> hm - damn vserver does not support mknod in the containers, this sucks
16:03 < rxr> and disqualifies them for building T2 within ... :-(
16:29 < CIA-8> rene * r26268 /trunk/package/www/owb/ (. compile.patch owb.cache owb.conf owb.desc): * added owb (1.0) - Origyn Web Browser designed for CE devices
16:34 -!- Dallur [n=earl@fgfw.basis.is] has quit [Read error: 110 (Connection timed out)]
16:37 -!- Key_Gena95 [i=lok@gateway/tor/x-fe2e366434bb8660] has joined #t2
16:48 < CIA-8> rene * r26269 /trunk/package/games/libkxl/libkxl.desc: * removed duplicate CROSS tak from libkxl
16:51 < CIA-8> rene * r26270 /trunk/package/emulators/hercules/hercules.desc: * removed duplicate CROSS tag from hercules, likewise
17:12 [Users #t2]
17:12 [@ChanServ] [ emte ] [ happycube ] [ mtr ] [ rxr ]
17:12 [ CIA-8 ] [ Enqlave] [ Key_Gena95] [ pdinoto_] [ sepp_ ]
17:12 [ cyda ] [ Fatal ] [ LMJ ] [ Ragnarin] [ valentin]
17:12 -!- Irssi: #t2: Total of 15 nicks [1 ops, 0 halfops, 0 voices, 14 normal]
17:20 < pdinoto_> hi rene!
17:20 < pdinoto_> just wanted to let you know that I am using my T2 "micro" package, which is based
17:21 < pdinoto_> on tresc. It is a tiny X system with networking and lua... boots in 5 secs
17:21 < pdinoto_> on a pentium board, runs on 18mb of ram (x with 4 xterms opened)
17:21 < rxr> nice :-)
17:21 < pdinoto_> and is under 32mb of disk uncompressed. Yay!
17:22 < pdinoto_> s/package/target/
17:22 < rxr> can you publish it so others can enjoy it, too ?
17:23 < pdinoto_> sure... it is a little bit messy (I keep playing with package list as I need them).
17:23 < pdinoto_> Do I send the files to your email?
17:25 < rxr> yes, or the t2 list: t2@t2-project.org
17:51 < CIA-8> rene * r26271 /trunk/package/graphic/tesseract/tesseract.desc: * updated tesseract (2.00 -> 2.01), added CV-URL tag
18:00 < rxr> Ragnarin: btw. the ide-generic is already loaded last, hm ?
18:00 < CIA-8> rene * r26272 /trunk/package/base/mkinitrd/mkinitrd.sh: * include chroot in the default initrd, just handy to be around
18:02 < CIA-8> mtr * r26273 /trunk/package/ (3 files in 2 dirs): * updated x264 (20061104-2245 -> 20071007-2245)
18:03 < CIA-8> mtr * r26274 /trunk/package/multimedia/ffmpeg/ (ffmpeg.desc libpostproc-destdir.patch): * updated ffmpeg (2007-04-25 -> 2007-10-07)
18:44 -!- Dallur [n=earl@fgfw.basis.is] has joined #t2
18:45 -!- Dallur [n=earl@fgfw.basis.is] has quit [Remote closed the connection]
19:34 -!- cgetty [n=Clark@pool-71-108-239-203.lsanca.dsl-w.verizon.net] has joined #t2
19:36 < Ragnarin> rxr: oh it is? hmmmm
19:37 < rxr> Ragnarin: I posted quite some stuff to you today ...
19:37 < rxr> did you got all scrollback ?
19:38 < Ragnarin> yea
19:38 < rxr> first of all did you notice the util-vserver is named utils-vserver in T2 ?
19:38 < rxr> is this useful for anything but global confusion =?
19:38 < rxr> can we fix this ?
19:38 < Ragnarin> more confusing is that we have vserver-utils and utils-vserver
19:39 < Ragnarin> we should
19:39 < rxr> yeah - but this vserver-utils is this new thing ...
19:39 < rxr> I think that naming is ok, as it is upstream ...
19:39 < rxr> do you use vserver ?
19:40 < Ragnarin> I try to ... trying to get a test system running
19:40 < rxr> what do you wanna use it for ?
19:41 < Ragnarin> user isolation
19:41 < rxr> ah
19:41 < rxr> I got a guest up today, works so far, _but_ vserver does not virtualize /dev/*
19:41 < Ragnarin> you can do that with grsec tho
19:41 < rxr> so it forbids mknod() so the guest containers can not create new ones
19:42 < rxr> oh
19:42 < Ragnarin> the guests shouldn't be able to create new dev nodes anyway?
19:42 < rxr> ok, I wanted to get openvz packaged for T2, now because it looks like opnevz also allows restricting device node access (and not just forbid mknod)
19:43 < rxr> actually I only need this lightweight containers to give balzius a partition of our new 8-core monser ...
19:43 < rxr> for t2/security work ...
19:43 < Ragnarin> ah ok :)
19:43 < rxr> so I need somehting where mknod and chroot work ...
19:43 < Ragnarin> chroot should work
19:43 < rxr> I package openvz now, we should have it anyway
19:43 < rxr> let's now how this turns out ...
19:44 < rxr> who renames this utils-verver? do you want to ?
19:44 < Ragnarin> do you know how? :)
19:44 < rxr> rename ?
19:45 < Ragnarin> can I do that on the whole dir?
19:45 < Ragnarin> without messing up the history?
19:45 < rxr> ah yes, files inside a dir and a dir in one svn commit was tricky
19:45 < rxr> I'll figure it out and post the comments "how to" to our svn page
19:46 < Ragnarin> ok good :)
19:47 < rxr> regarding the ide-generic, it is after the udevstart / udevtrigger thing
19:47 < rxr> it should be loaded after the udev stuff
19:47 < Ragnarin> btw does dev.croup.de resolve for you?
19:47 < rxr> though - maybe - we should insert udevsettle as barrier so it is not loaded quicker than the udev even processing takes ...
19:47 < Ragnarin> are you sure the autodetection doesn't load it earlier?
19:48 < rxr> oh - you mean tha auto-detection loads generic-ide ?
19:48 < rxr> that could be, yes
19:48 < rxr> Unknown host dev.croup.de
19:48 < Ragnarin> I think so
19:49 -!- cgetty [n=Clark@pool-71-108-239-203.lsanca.dsl-w.verizon.net] has quit ["Leaving"]
19:50 < Ragnarin> hmm so vserver-utils is dead ... *googles*
19:51 -!- kic- [i=rfoote@208.115.42.120] has joined #t2
19:51 < kic-> re
19:51 < kic-> rxr leave already?
19:52 < rxr> nope
19:52 < Ragnarin> no
19:52 < rxr> hie
19:52 < kic-> wee
19:52 < kic-> heh
19:52 < kic-> hi
19:52 < rxr> still fighting virtual server crap
19:52 < kic-> ah
19:52 < Ragnarin> lol
19:52 < rxr> did you test the new image s?
19:52 < kic-> yanno those changes you made to the livecd/init that you emailed me about a little bit ago?
19:52 < kic-> not yet
19:53 < rxr> hm ?
19:53 < rxr> your second last post was too crypted for me .-)
19:54 < CIA-8> rene * r26275 /trunk/package/security/ (8 files in 2 dirs):
19:54 < CIA-8> * renamed utils-vserver -> util-vserver, as that is the upstream
19:54 < CIA-8> name and everything else is just confusing
19:54 < kic-> he changed his mind, and said he wouldn't mind waiting until you were able to get a good solution for it
19:54 < kic-> if it only took a day or two
19:54 < rxr> kic-: but the quickfix should be ready to test
19:55 < rxr> kic-: I could not test, as I can not copy the flash intended files over for quikc check, but I hope it works, and it is done now anyway ...
19:55 < rxr> it just needs a test run (and maybe a quick-fix interation in the case I made a typo :... )
19:56 < kic-> right..
19:57 < kic-> how difficult though is it to revert back?
19:57 < kic-> if the test does not go well?
19:57 < rxr> just a single file that is not even part of your target
19:57 < rxr> it turned out to be a two-liner
19:57 < kic-> ah
19:57 < kic-> nice
19:58 < kic-> did you back up old file? or hash out or something?
19:58 < rxr> but as it is not part of your target it's a little bad for versioning
19:58 < rxr> svn revert on the file will do :-)
19:59 < kic-> ugh
19:59 < kic-> hehe
19:59 < kic-> which file?
20:00 < rxr> target/share/livecd/init2
20:02 < rxr> ouh - damn
20:02 < rxr> something conflicts with the openvz patch, and I fear it's our t2 no2mod stuff
20:02 < rxr> damnit
20:02 < rxr> t
20:02 -!- mqueiros [n=mqueiros@80.243.90.138] has joined #t2
20:03 < rxr> ieeek
20:03 < rxr> they patch-in a don_ask mode ??!?!?!?!
20:03 < rxr> damn pseudo open source companies
20:03 < rxr> I know why I tried vserver before openvz ...
20:04 < rxr> I wonder if we need our no2mod patch at all - maybe it would just work if we yes 'm' | make oldconfig or so ..
20:09 < rxr> too bad atop conficts with openvz, likewise
20:13 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has joined #t2
20:16 -!- mqueiros [n=mqueiros@80.243.90.138] has quit []
20:16 < tri> moin
20:17 < rxr> moin tri
20:20 -!- Enq|xps [n=stealth@unaffiliated/enqlave] has joined #t2
20:26 < kic-> hrm, new jffs2 version out
20:27 < rxr> :-)
20:27 < rxr> you wanna patch it in ?
20:27 < kic-> would be wonderful if it will solve any of my headaches :D
20:28 < kic-> is that a kernel patch?
20:28 < tri> hi rene any news about the xen support in t2 ?
20:28 < tri> t2 as dom0 ?
20:28 < rxr> tri: with some manual work here and there yet
20:28 < tri> manuel work ?
20:28 < rxr> but as my interested in Xen dropped before I finished integration patches welcome ...
20:29 < tri> last time i wasn't able to build it :-)
20:29 < rxr> you have to build the kernel manually ...
20:29 < tri> why was your interested in xen dropped ?
20:29 < rxr> not dropped
20:29 < rxr> just not finished
20:29 < tri> :-)
20:30 < rxr> I'm on Linus and Alans side on this: an additional, crappy scheduler below linux just doesn't cut
20:30 < rxr> it
20:30 < rxr> runing the containers as Linux processes is more viable ..
20:30 < tri> i have the change to test xen on a big machine (quadcore 16gb ram) on wednesday
20:31 < tri> ah ok does kvm or something else work different there ?
20:31 < tri> or what do you mean as "containers" ?
20:32 < rxr> yes, kvm
20:32 < rxr> another thing you should consider: the fat xen machines have fixed memory patition as there are full OS instances in there
20:32 < rxr> tri: when you just have linux servers, using vserver or openvz saves you a lot of memory as the memory is dynamically allocated between all your virtual containers
20:33 < rxr> with something like xen / vmware you waste a lot of memory not only on the kernel's and their environment (library loaded etc.) but aos each VM will have free memory that is not shared among them
20:37 < tri> hm but with vserver i don't have a change to run win / netware and linux on one hardware
20:37 < tri> how about kvm
20:37 < tri> do you thing the idea behind kvm is better then the xen idea ?
20:38 < rxr> well - atleast for the "reusing the Linux process scheduler and memory management"
20:38 < rxr> it has less management tools around, though - however, I don't like the Xen python bloat ... so that might be a + :-)
20:38 < rxr> maybe check them all out
20:38 < tri> :-)
20:38 < rxr> and if you can finish Xen integration in T2 :-)
20:39 < tri> hm yeah i will try to use t2 - but i think i have to use novell linux - "standard company linux"
20:40 < tri> :-(
20:41 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has left #t2 []
20:41 -!- pdinoto_ [n=pdinoto@200.117.246.103] has quit [Read error: 104 (Connection reset by peer)]
20:41 -!- pdinoto_ [n=pdinoto@200.117.246.103] has joined #t2
20:46 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has joined #t2
20:48 < tri> re
20:48 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has left #t2 []
20:48 < rxr> damnit
20:48 < rxr> at least the vserver code compiled out of the box ...
20:48 * rxr slabs this rusian SWSoft el-cheapo coders ...
20:50 < rxr> 2.6.22 ovz kernel is in early alpha, and chechpointing functionality is
20:50 < rxr> not yet fully ported. For now, you have to deselect it if you want to
20:50 < rxr> try compiling that kernel.
20:50 < rxr> hm
20:54 -!- mqueiros_ [n=mqueiros@80.243.90.138] has joined #t2
21:10 < rxr> btw:
21:10 < rxr> http://people.exactcode.de/~rene/_G0D6747.jpg
21:10 < rxr> http://people.exactcode.de/~rene/_G0D6749.jpg
21:12 [Users #t2]
21:12 [@ChanServ] [ emte ] [ Fatal ] [ kic- ] [ mtr ] [ rxr ]
21:12 [ CIA-8 ] [ Enqlave] [ happycube ] [ LMJ ] [ pdinoto_] [ sepp_ ]
21:12 [ cyda ] [ Enq|xps] [ Key_Gena95] [ mqueiros_] [ Ragnarin] [ valentin]
21:12 -!- Irssi: #t2: Total of 18 nicks [1 ops, 0 halfops, 0 voices, 17 normal]
21:17 < sepp_> i c red
21:28 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has joined #t2
21:43 < LMJ> Nice
21:45 < tri> re
21:46 < LMJ> NICE
21:49 < tri> what is nice ?
21:50 < mqueiros_> tri: I believe LMJ was talking about this: http://people.exactcode.de/~rene/_G0D6747.jpg and http://people.exactcode.de/~rene/_G0D6749.jpg
21:50 < LMJ> ;)
21:52 < tri> cool :-)
21:52 < tri> my girlfriend
21:55 < kic-> thats not your gf
21:55 < kic-> :D
21:56 < tri> ah ok, right :-)
21:56 < kic-> is it?
21:56 < kic-> lol
21:56 < tri> do you know that shirt "you need a girfriend without *.jpg in here name"
21:57 < kic-> no
21:57 < kic-> they make a shirt that says that?
21:57 < kic-> i want one
21:57 < kic-> heh
21:57 < tri> :-)
21:57 < kic-> omg, i know i'm looking at to much code when i browse some C code for jffs2 patch in the kernel, and i begin to understand what it means =(
21:58 < kic-> apparently god is going to force me to learn C one way or another
21:58 < tri> hm how many mem can i allocat at one time under linux ? (x86)
21:59 < kic-> *shrug* the full amount of your ram chip?
21:59 < kic-> heh
21:59 < kic-> can't you vmalloc the whole thing?
21:59 < kic-> if you wanted to :D
22:04 < Ragnarin> there's a ulimit setting usually
22:05 < tri> hm the ulimit setting on my box is unlimited
22:05 < tri> can i allocate as many i like to ?
22:05 < Ragnarin> then you'd be limited by the address space you get per process
22:06 < Ragnarin> which can be limiting on 32bit at least
22:06 < tri> what happens if i allocate "to much" - is this possible
22:07 < tri> i have read something about oom killer
22:07 < Ragnarin> well look at for example the ace package configure script (among others)
22:08 < Ragnarin> when the machine runs out of virtual memory (real memory + swap) some processes will be killed
22:10 < tri> hm http://kerneltrap.org/node/1010
22:28 -!- TobiX [i=tobias@zoidberg.org] has joined #t2
22:29 < Ragnarin> seems the problem there is that the kernel gives out memory that's not actually there ... that's bad
22:50 < tri> gn8
22:50 -!- tri [n=tri@p57ADA139.dip0.t-ipconnect.de] has left #t2 []
--- Log closed Tue Oct 09 00:00:42 2007