--- Day changed

--- Log opened Sun Jun 29 00:00:02 2003
00:20 -!- kayl [~jada@6532176hfc248.tampabay.rr.com] has quit []
00:40 -!- umnode1 [geeks@ool-44c02046.dyn.optonline.net] has quit [Read error: Connection reset by peer]
01:49 < lah> oh, pflanze also wanted skas+grsec :)
02:06 -!- albino [~albino@neuf.dorms.usu.edu] has quit [Ping timeout: 496 seconds]
02:10 -!- wind [hscbo@61.149.7.226] has joined #uml
02:10 < wind> hi all
02:11 < wind> i wonder whether two uml can communication without router on a same host
02:11 < wind> help me
02:12 -!- splinter [~peatah@210.49.137.83] has joined #uml
02:16 < wind> who can help me?
02:16 < Pahan> wind: Why don't you read the "virtual networking" section on the UML website?
02:16 < Pahan> wind: It describes everything you need.
02:16 < wind> thanks
02:17 < wind> but could you like answer my question?
02:17 < splinter> has anyone here tried out linode.com?
02:17 < caker> yeah - they rock
02:18 < wind> cause i have no time
02:18 < Pahan> wind: Well, then don't waste our time.
02:18 < caker> wind: use a bridge with tuntap devices on the host -- go read that section
02:19 < wind> i tried tuntap,but uml cannot access host
02:20 < wind> pahan: sorry, but i don't think i wasted your time.
02:20 < Pahan> wind: I tried to conserve it by not describing things you are perfectly capable of doing on your own.
02:21 < wind> i have the right to ask question, and you have rigth to keep slient.
02:21 < wind> sorry
02:22 < Pahan> oh wtf
02:22 < Pahan> Are you from USA or something?
02:22 < splinter> lol caker, i think you're kind of biased :-)
02:22 < caker> splinter: well, I can have an opinion :-)
02:23 < caker> heh
02:24 < Pahan> "Each machine is split differently to guarantee your CMR (committed MHZ rate)"
02:24 < Pahan> wtf is that?
02:26  * lah is at rh9 install once again (for the 500th time)
02:32 -!- wind [hscbo@61.149.7.226] has quit []
02:33 < Pahan> Hrm, how does linode migrate UMLs around?
02:35 < caker> they don't migrate unless you've got shared storage
02:36 < Pahan> How can you migrate them if you do? What is "shared storage", some arcane fiber stuff?
02:36 < caker> it could be as simple as an nfs volume
02:36 < caker> over ethernet
02:36 < caker> It would just be from boot to boot, no migration while running .. yet :-)
02:36 < Pahan> Having to shut them down doesn't count.
02:37 < caker> counts for 100% redundancy
02:38 < lah> yeeee!
02:39 < lah> it finished the installation without commiting suicide (hanging) ;)
02:39 < caker> lah: all right .. only 501, huh?
02:39 < lah> i think cleaning up the kernel ("ptrace failed" blah) helped a lot ;)
02:40 < lah> caker: yeah, it kept breaking at "performing post isntall.." while executing rpm somehow
02:40 < lah> caker: but worked after cleaning the kernel
02:44 < Pahan> caker: So, what is the linode setup fee?
02:49 < splinter> and for how long is there no setup fee?
02:49 < lah> *fixing /dev*
03:09 < Dave\\> does linode provide their rootfs's for free to the uml community and a mirror like it says on their site?
03:09  * Dave\\ needs stable rootfs badly
03:09 < Dave\\> lol
03:09 < caker> Dave\\: not yet
03:10 < Dave\\> ok
03:11 < Dave\\> if I sign up for linode, do I get access to the images or is that all handled by the automation?
03:11 < caker> automated - but you'd be inside the filesystem itself...
03:12 < splinter> wow! i signed up 5 minutes ago and already i'm in my deb system :)
03:15 -!- ElectricElf [david@elf.noc.oftc.net] has quit []
03:46 -!- ElectricElf [david@elf.noc.oftc.net] has joined #uml
03:57 -!- albino [~albino@neuf.dorms.usu.edu] has joined #uml
03:58 < ido> eh. sweet. 
03:58 < ido> how's the speed at linmode ?
04:00 < splinter> haven't tested it yet
04:00 < splinter> but it's 20 hops from my isp :(
04:03 < ido> is it down now or is it just me ?
04:04 < splinter> just you :)
04:48 -!- revenger2 [revenger@p508B2D11.dip.t-dialin.net] has joined #uml
04:50 -!- revenger_ [revenger@p508B3973.dip.t-dialin.net] has quit [Ping timeout: 492 seconds]
07:15 -!- DnsInfector [~DnsInfect@ARennes-204-1-12-29.w81-53.abo.wanadoo.fr] has joined #uml
07:45 -!- Lathiat [lathiat@seven.sixlabs.org] has quit [Quit: reboot]
07:49 -!- Lathiat [lathiat@seven.sixlabs.org] has joined #uml
10:34 -!- Lathiat [lathiat@seven.sixlabs.org] has quit [Quit: Lost terminal]
10:35 -!- Lathiat [lathiat@130.95.13.25] has joined #uml
10:43 -!- ElectricElf [david@elf.noc.oftc.net] has quit [Ping timeout: 245 seconds]
11:56 -!- ASY [~someone@216.221.51.240] has quit [Ping timeout: 488 seconds]
11:58 -!- shak [~shak@80.4.84.78] has joined #uml
12:01 < shak> hrm
12:01 < shak> I've made a SuSE filesystem
12:01 < shak> but, Im having a problem..
12:01 < shak> it gets to " Starting hardware scan on boot" and locks :/
13:14 < shak> hrm
13:14 < shak> anyone alive?
13:54 < lah> no!
13:54 < lah> :)
14:53 < shak> hrm
14:53 < shak> the networking options have dissappeared from make menuconfig
14:53 < shak> :/
15:14 < caker> for a UML kernel?
15:14 < shak> no
15:14 < shak> I've fixed it now, it was menuconfig being silly
15:14 < shak> hadn't registered that I'd done general-> networking
15:15 < shak> so I just did make mrproper and make clean again
15:15 < shak> and it was ok
15:26 -!- trappist [~trappist@4.47.221.175] has joined #uml
15:27 < trappist> is there any hope of uml/skas being merged?  has linus or any of the other maintainers commented on it?
15:28 < silug> they don't want to touch the /proc/mm stuff...
15:28 < trappist> I know skas messes with that, does uml?
15:28 < silug> jdike needs to make everything syscalls, then they might go for it
15:29 < trappist> do the -ac kernels still have uml?
15:30 < silug> it's probably not up-to-date
15:30 < caker> trappist: UML is included in 2.5, although Jeff has said Linux hasn't pulled his changes for a *long* time
15:30 < caker> trappist: regardless, skas isn't getting into main
15:30 < trappist> that's what I was looking for :(
15:30 < trappist> thanks
15:30 < caker> ack - i just said Linux instead of Linux :-)
15:31 < trappist> linus?
15:31 < trappist> :)
15:31 < caker> damnit, i did it again .. Linus. that is hard to write
15:31  * caker is too tired
15:32 < caker> now with Linus working full-time on the kernel (has he ever not?) perhaps we'll see more frequent pulls ...
15:32 < trappist> that would be nice
15:33 < trappist> he doesn't touch 2.4.x at all though does he?
15:34 < trappist> I sure wouldn't mind seeing uml in 2.4
15:34 < trappist> hacking those patches into mandrake kernels is tedious :/
15:35 < caker> why bother? what does the md kernel have going for it?
15:35 < trappist> lots of bloat :)
15:35 < trappist> I'm on the secteam and I want mandrake kernels with all the grsecurity goodies etc. so I can set up a valid test environment
15:36 < caker> ahh
15:40 -!- ElectricElf [david@elf.noc.oftc.net] has joined #uml
15:43 < david> hello
15:43 < caker> afternoon
15:43 < david> trappist: grsecurity hasn't been ported to arch-um - It's not going to work right
15:44 < caker> I think he means grsecurity and skas
15:44 < david> oh
15:44 < david> ok
15:45 < caker> which hacking is still required to make work, right?
15:45 -!- rawl [~rawl@cable200-75-79-20.epm.net.co] has joined #uml
15:45 < david> right
15:46 < caker> thought so
15:46 < david> grsecurity stops ptrace from working, I believe
15:51 < shak> do does the ptrace patch
15:52 < david> shak: er, that ptrace patch is broken anyway
15:52 < david> shak: 2.4.21 has a much improved ptrace fix
15:52 < shak> Im running .21
15:52 < shak> the ptrace patch stopped me from exploiting my own box
15:52 < david> ok, and?
15:53 < shak> so it can't be that bad.
15:53 < david> I have no idea what you're talking about
15:53 < david> stopping ptrace makes UML not work
15:53 < david> the ptrace patch just stops non-root users tracing certain kernel threads
15:53 < shak> dof
15:53 < shak> doh, sorry 
15:53 < shak> misread
15:54 < shak> ptrace root problem.
15:54 < shak> you could root the box via ptrace
15:54 < shak> that's what Im talking about
15:54 < david> oh, right
15:54 < david> that was fixed like 3 months ago :-)
15:55 < shak> yes
15:55 < trappist> hum
15:55 < shak> umount: /tmp/target/proc: not found
15:55 < shak> warning: can't open /etc/fstab: No such file or directory
15:55 < shak> mount: can't find / in /etc/fstab or /etc/mtab
15:55 < shak> System halted.
15:55 < shak> hrm
15:55 < shak> sorry, didnt mean to paste that much
15:55 < shak> rootstrap doesnt like me
15:55 < trappist> grsec is compiled out of the default kernel but I was kinda hoping to test kernel-secure, which has grsec... ain't gonna work, eh?
15:56 < trappist> rootstrap?
15:56 < david> trappist: if it's i386 specific then it's not going to work
15:56 < shak> to create the rootfs
15:57 < trappist> haven't heard of that one
15:57 < trappist> I've sure been looking for a howto to make one from scratch
15:57 < trappist> I used umlbuilder which is nice but I don't have as much control as I'd like
16:01 < mistik1> hey guys
16:02 < trappist> heya mistik1
16:02 < mistik1> trappist: out for some more fun, eh?
16:02 < trappist> yeah I kinda forgot how much fun this can be :)
16:02 < mistik1> I need to upgrade my utils and kernel
16:03 < mistik1> I used umlbuilder to setup my filesystems but I hacked the living hell out thier control script
16:03 < mistik1> and removed thier method of networking on the UM side
16:04 < mistik1> I dislike uml_net and also do not want to use IPs on both sides
16:05 < trappist> not getting any non-debian hits on a google search for rootstrap, should I take this as a hint?
16:05 < mistik1> haha
16:08 < mistik1> trappist: what you trying to get done?
16:08 < trappist> mistik1: mostly I'd like to be able to have a lot more control over the building of the filesystem so I can guarantee its integrity vs. a real install
16:09 < mistik1> caker and I were talking about the need for a sort of super uml builder/runner
16:09 -!- shak [~shak@80.4.84.78] has quit [Read error: Connection reset by peer]
16:11 < mistik1> a very granular tool for building UM filesystems(even on the fly) and taking care of the network setup on host among whatever else we can think of
16:13 < silug> it would be nice if uml in the stock kernel was reasonably current.  then maybe the distributions would work on making it easier to do installs under uml.
16:13 < silug> although they probably wouldn't
16:15 < silug> debian might at least though...
16:19 -!- shak [~shak@pc1-hudd4-5-cust78.hudd.cable.ntl.com] has joined #uml
16:19 < shak> hmm, does anyone else notice that the UMLs load artificially high?
16:20 < shak> root@uml1:~# uptime
16:20 < shak>  20:18:38 up 2 min,  2 users,  load average: 2.62, 1.02, 0.37
16:30 < caker> trappist: try debootstrap
16:31 < shak>  20:29:26 up 12 min,  2 users,  load average: 3.00, 2.75, 1.63
16:31 < shak> hrm
16:31 < shak> root@uml1:/var# vmstat
16:31 < shak>    procs                      memory    swap          io     system         cpu
16:31 < shak>  r  b  w   swpd   free   buff  cache  si  so    bi    bo   in    cs  us  sy  id
16:31 < shak>  0  3  0      0 110216    812   8728   0   0    22     2  104    12   1   0  99
16:31 < shak> its not got any CPU usage.
16:31 < shak> brb
16:31 < shak> going to reboot the host
16:31 -!- shak [~shak@pc1-hudd4-5-cust78.hudd.cable.ntl.com] has quit [Quit: Client exiting]
16:55 -!- lah [~z@AVATAR.internet.ufg.ac.at] has quit [uranium.oftc.net arion.oftc.net]
16:59 -!- lah [~z@AVATAR.internet.ufg.ac.at] has joined #uml
17:08 < david> hello
17:08 -!- trappist [~trappist@4.47.221.175] has quit [Ping timeout: 488 seconds]
17:40 < ido> if i assign a file a size, and mkfs it, could i enlarge it later on in the future ?
17:46 < Pahan> Yes, use resize2fs or however it is called.
17:47 < Pahan> You can seek beyond the EOF and write a byte, then close the file and use the resize thingie.
17:49 < ido> oh, i see
17:49 < Pahan> You can fuck with dd, or you can use Python interactive mode ;-)
17:50 < ido> whats python interactive mode ?
17:51 < Pahan> Python interpreter can be used interactively.
17:51 < ido> for what ?
17:51 < Pahan> To extend file size, for example.
17:51 < Pahan> The simplest way to do it.
17:51 < ido> oh
17:51 < ido> i see.
17:51 < ido> don't have a clue how though.
17:52 < Pahan> Do you need to do it now?
17:52 < ido> no.
17:52 < Pahan> :P
17:53 < ido> damnit, something happend yesterday, the UML crashed
17:53 < ido> and since then , i cant seem to be able to boot up any UML
17:53 < Pahan> "UML crashes so nicely, NOBODY NOTICES"
17:53 < Pahan> Oh.
17:54 < ido> eh
17:54 < Pahan> Never mind.
17:54 < ido> i mean, it boots alright, but doesn't give me a VC
17:55 < Pahan> Then you haven't specified your consoles right.
17:55 < ido> what do you mean ?
17:55 < Pahan> Perhaps they are popping on xterms in an unexpected display.
17:56 < ido> i run it with con=null con0=fd:0,fd:1
17:56 < Pahan> Heh, null.
17:56 < ido> yep
17:56 < Pahan> That certainly ensures you'll never find out what the problem is.
17:57 < ido> how can i check whats wrong ?
17:57 < Pahan> By not redirecting con to null.
17:57 < ido> i mean, even if i boot it without and con* options, i still get the same thing
17:57 < Pahan> Oh, wiat.
17:58 < Pahan> You disabled your VCs with "con=null"
17:58 < Pahan> con0 is the boot messages.
17:58 < ido> aha
18:00 < ido> it gets stuck right after :  * Starting local...  [ ok ]
18:00 < Pahan> Heh, RedHat.
18:00 < Pahan> I don't want to know what it does.
18:00 < ido> gentoo actually
18:01 < ido> Suddenly the Dungeon collapses!! - You die...
18:01 < ido> wtf ?
18:02 < Pahan> Got hax0rs?
18:02 < Pahan> Or some pindick from gentoo decided to make a funney?
18:02 < ido> no, that happend after i killed the UML
18:05 < ido> hmm
18:05 < ido> i do seem to get virtual consoles in X
18:06 < ido> although i dont get the 'terminal' virtual console
18:21 < silug> ido: i think that message is from screen.
18:21 < silug> $ strings /usr/bin/screen | grep Dungeon
18:21 < silug> Suddenly the Dungeon collapses!! - You die...
18:22 < silug> i thought i had seen that before.  i'm not sure why.
18:22 < ido> eh, apparently it is...
18:23 < ido> i still have no idea why whats wrong with the virtual terminal
18:28 < silug> try "reset".  maybe something is wrong with your terminal.
18:28  * silug makes a wild guess
18:32 < ido> could be
18:41 -!- ido [~Ido@bzq-218-140-160.cablep.bezeqint.net] has quit [Ping timeout: 496 seconds]
19:13 -!- DnsInfector_ [~DnsInfect@ARennes-204-1-2-17.w193-251.abo.wanadoo.fr] has joined #uml
19:18 -!- DnsInfector [~DnsInfect@ARennes-204-1-12-29.w81-53.abo.wanadoo.fr] has quit [Ping timeout: 490 seconds]
20:51 < david> boo
20:57 -!- Dave\\ [dave@67.81.154.64] has quit [Ping timeout: 488 seconds]
21:07 -!- splinter [~peatah@210.49.137.83] has left #uml []
21:15 < gump> m00
21:29 -!- Dave\\ [dave@67.82.225.53] has joined #uml
22:24 -!- rawl [~rawl@cable200-75-79-20.epm.net.co] has quit [Quit: Client exiting]
22:56 < david> caker: around?
22:58 -!- DnsInfector_ [~DnsInfect@ARennes-204-1-2-17.w193-251.abo.wanadoo.fr] has left #uml [Client exiting]
23:42 < caker> david: yes
--- Log closed Mon Jun 30 00:00:00 2003