Back to Home / #uml / 2007 / 12 / Prev Day | Next Day
#uml IRC Logs for 2007-12-12

---Logopened Wed Dec 12 00:00:13 2007
00:00|-|hfb [~hfb@75.80.37.175] has joined #uml
00:08|-|balbir [~balbir@122.167.208.28] has joined #uml
01:54|-|balbir [~balbir@122.167.208.28] has quit [Ping timeout: 480 seconds]
02:06|-|balbir [~balbir@122.167.179.91] has joined #uml
03:20<ferret_0567-#uml->>does anyone here know how to fix a iptables problem I'm having? Port 345 won't forward to WAN...
03:20<ferret_0567-#uml->>iptables -t nat -A prerouting_wan -p tcp --dport 80 -j DNAT --to 192.168.1.103
03:20<ferret_0567-#uml->>iptables -A forwarding_wan -p tcp --dport 80 -d 192.168.1.103 -j ACCEPT
03:23<ferret_0567-#uml->>nvm
05:19|-|krau [~cktakahas@200.184.118.132] has quit [Ping timeout: 480 seconds]
05:37|-|fo0bar [fo0bar@feh.colobox.com] has quit [Ping timeout: 480 seconds]
05:37|-|fo0bar [fo0bar@feh.colobox.com] has joined #uml
05:38|-|namit [~test_acco@83-70-39-95.b-ras1.prp.dublin.eircom.net] has joined #uml
05:42|-|Hunger [Hunger.hu@Hunger.hu] has joined #uml
06:11|-|krau [~cktakahas@200.184.118.132] has joined #uml
06:51|-|karol [~karol@chello089076073248.chello.pl] has joined #uml
06:52|-|karol changed nick to Magotari
08:19|-|IntuitiveNipple [~TJ@alexandros.tjworld.net] has joined #uml
08:45|-|Infinito [argos@200-140-154-162.gnace701.dsl.brasiltelecom.net.br] has joined #uml
09:02|-|dang [~dang@nemesis.fprintf.net] has quit [Ping timeout: 480 seconds]
09:30|-|dang [~dang@aa-redwall.nexthop.com] has joined #uml
10:02|-|cl4sh [~cl4sh@qik.ds.pg.gda.pl] has joined #uml
10:02<cl4sh-#uml->>hi all
10:06<Magotari-#uml->>Hey, cl4sh.
10:07<cl4sh-#uml->>ive got new quest i want to prevail my promotor of mos work to subject-matter
10:07<cl4sh-#uml->>connected with uml/vnum;
10:07<cl4sh-#uml->>vnuml
10:07<Magotari-#uml->>Heh.
10:08<Magotari-#uml->>This will hurt, I imagine.
10:08<cl4sh-#uml->>Master OF science promotor
10:08<cl4sh-#uml->>no
10:08<cl4sh-#uml->>he wants two pages descriptions of vnuml/uml
10:08<Magotari-#uml->>If you have any question as to the UML side of things, feel free to ask. Jeff should be here shortly too.
10:09<cl4sh-#uml->>uhm
10:09<Magotari-#uml->>Hmm...
10:09<cl4sh-#uml->>first
10:09<cl4sh-#uml->>i need good art about things that i can do under uml
10:09<Magotari-#uml->>I think I saw a link about that. Sec.
10:09<cl4sh-#uml->>esspecialy connected with network
10:09<cl4sh-#uml->>or telecommunications
10:10<jcpiza-#uml->>hi cl4sh
10:10<cl4sh-#uml->>hi
10:10<jcpiza-#uml->>i've a question about vnuml
10:11<cl4sh-#uml->>yes?
10:11<Magotari-#uml->>http://user-mode-linux.sourceforge.net/old/uses.html
10:11<jcpiza-#uml->>what patches i need? kernel + um + ...?
10:11<Magotari-#uml->>jcpiza: You need mconsole exec too, I think.
10:11<jcpiza-#uml->>kernel vanilla
10:11<cl4sh-#uml->>which distro?
10:11<jcpiza-#uml->>usermodelinux
10:11<jcpiza-#uml->>and a lot of patches unknown for me
10:12<jcpiza-#uml->>slackware
10:12<dang-#uml->>cl4sh: Any distro can run UML.
10:12<cl4sh-#uml->>one moment
10:12<cl4sh-#uml->>dang: yes
10:12<dang-#uml->>It's part of the vanilla kernel.
10:12<Magotari-#uml->>Yeah, but vnuml ain't a nice toy.
10:12<cl4sh-#uml->>dang: but theres howto about uml for few distros
10:12<Magotari-#uml->>You need a special kernel for some things.
10:12<cl4sh-#uml->>tfu vnuml
10:12<cl4sh-#uml->>yes You need it
10:12<cl4sh-#uml->>but last time
10:13<cl4sh-#uml->>ive run simulation scenario with kernels from nnagafix
10:13<cl4sh-#uml->>and it works fine
10:14<jcpiza-#uml->>only modern CPUs (with hw hypervisor HVM) have benefits that have not the old CPUs, see http://unit.aist.go.jp/itri/knoppix/vmknoppix/index-en.html
10:14<jcpiza-#uml->>and http://unit.aist.go.jp/itri/knoppix/xen/index-en.html made by japanese ninja
10:15<dang-#uml->>Meh. I wrote my own scripts that seem to be as good as VNUML
10:15<cl4sh-#uml->>:)
10:16<cl4sh-#uml->>im not good with this things
10:16<cl4sh-#uml->>but seems that i should be
10:16<Magotari-#uml->>Hmm...
10:16<Magotari-#uml->>I don't think you can use any special hw with um.
10:16<Magotari-#uml->>uml, I mean.
10:16<jcpiza-#uml->>the minimum CPU that i've to buy is Core Duo 2 or modern Athlon64 with Pacifica
10:17<cl4sh-#uml->>and in case of my mos work the vnuml is only an example of toot to simulate net
10:17<Magotari-#uml->>UML is paravirtualization with the host kernel being the hypervisor for the guest. No KVM or anything else support as of yet...
10:17<jcpiza-#uml->>so i could simulate windows and linux scenarios
10:18<cl4sh-#uml->>only linux i guess
10:18<jcpiza-#uml->>cl4sh, no *bsd?
10:19<cl4sh-#uml->>hmm idont know
10:19<jcpiza-#uml->>UML has limitations
10:19<Magotari-#uml->>Yes, but benefits too.
10:20<Magotari-#uml->>Speed is pretty nice, so is ease of use. And running a real distro on a real partition is easy too.
10:20<cl4sh-#uml->>as i look at the vnuml docs it could be ten nodes and with kernel 2.6 about 64 networks
10:21<cl4sh-#uml->>but whos got so many memory on his machine :P
10:22<cl4sh-#uml->>Magotari: i can say that i was simulate two asterisk central and connect it between themselves
10:22<cl4sh-#uml->>it was work really fine
10:22<Magotari-#uml->>That is pretty good.
10:22<jcpiza-#uml->>i will put 4 modules of 2 gibibytes to a modern PC, 8 gebibytes is sufficient.
10:23<Magotari-#uml->>If you need memory, you can always get swap. It works.
10:23<Magotari-#uml->>I could run a 3GB UML right now, and I have 1GB of ram.
10:23<jcpiza-#uml->>Magotari, swap is superslow!
10:23<Magotari-#uml->>Yeah, but it works if you need it.
10:23<cl4sh-#uml->>Magotari: thx for arts maybe i'll back here in the evning
10:23<Magotari-#uml->>Right. See you later then.
10:23<jcpiza-#uml->>55 MB/s flushing to/from disk is bad idea
10:24|-|cl4sh [~cl4sh@qik.ds.pg.gda.pl] has left #uml [uml]
10:24<jcpiza-#uml->>i need GB/s!
10:24<Magotari-#uml->>Swap is cheaper, and teaches you patience.
10:25<Magotari-#uml->>:)
10:25<jcpiza-#uml->>it's unpatiencely horrourous
10:25<Magotari-#uml->>Heh. I wish I could try a fully swap backed UML now.
10:26<jcpiza-#uml->>works UML in modern PC x86-64 8 GiB ram from shop?
10:26<dang-#uml->>Um, yes. :)
10:26<Magotari-#uml->>Never tried it in those conditions. It does have some highmem support though. I think it should work.
10:26<dang-#uml->>UML works on anything. All that changes is the speed...
10:26<jcpiza-#uml->>i can simulate a city-sized network (200 nodes)
10:26<dang-#uml->>Yes.
10:27<dang-#uml->>I've done 250 on my machine.
10:27<dang-#uml->>It's slow, tho, on this box.
10:27<Magotari-#uml->>You can, but maybe Operating system virtualization could be a better choice...
10:27<jcpiza-#uml->>dang, superslow
10:27<dang-#uml->>Magotari: Not to simulate networks...
10:27<Magotari-#uml->>Take a look at OpenVZ.
10:27<Magotari-#uml->>Oh. You can't do that, eh? Sorry then. I was not aware.
10:27<Magotari-#uml->>I hate networking. :/
10:27<dang-#uml->>I haven't found anything to beat UML for simulating complex networks...
10:28<dang-#uml->>All the virtualization stuff I've tried just does a single virtual network kind of thing.
10:28<dang-#uml->>With UML, you can build arbitrary topologies, easily.
10:28<Magotari-#uml->>qemu seems to have some support... Not as nice, but has it.
10:28<jcpiza-#uml->>they use virtual slow pings, not true pings
10:29<Magotari-#uml->>Hmm...
10:29<Magotari-#uml->>Qemu is compatible with mcast of UML?
10:29<Magotari-#uml->>That would be pretty nice. OpenBSD and UML networked together...
10:29<jcpiza-#uml->>Hahaha
10:29<dang-#uml->>This is the drawback of UML, of course. Only linux.
10:30<dang-#uml->>Magotari: I'm pretty sure qemu can connect to a tun/tap
10:30<dang-#uml->>And so can UML.
10:30<Magotari-#uml->>Yes, it can.
10:30<jcpiza-#uml->>qemu can use tuntap to interact *BSD with linux virtualboxes
10:30<Magotari-#uml->>Yes, but that in tuntap.
10:30<Magotari-#uml->>I am talking about mcast.
10:30<Magotari-#uml->>Which it seems to also have.
10:30<Magotari-#uml->>Not like I tested it, but... The man page said so.
10:31<jcpiza-#uml->>magotari, it's independent of the protocol
10:31<dang-#uml->>Seems unlikely...
10:31<jcpiza-#uml->>you can configure qemu to use one IP and one MAC
10:33<jcpiza-#uml->>qemu is slow, the most complex project is to build VNUML + XEN + VIRTUALBOX + KQEMU +
10:33<jcpiza-#uml->>etc.
10:34<jcpiza-#uml->>e.g., i want to run 3 windows ( 98, 2000, xp ), 2 bsd ( fbsd and obsd ) and 100 linuxes
10:35<Magotari-#uml->>I'd stick to just (Xen or Kqemu) and uml in this scenario. Why add the complexity of Vbox and (Xen or Kqemu)?
10:36<Magotari-#uml->>Xen and Kqemu let you run an arbitrary os, both of them.
10:36<Magotari-#uml->>Why use both technologies?
10:38<jcpiza-#uml->>Magotari, http://unit.aist.go.jp/itri/knoppix/xen/index-en.html to see Performance, qemu(kqemu) 24.87s vs Xen(DomU) 14.68s
10:39<jcpiza-#uml->>Xen(HVM) 15.99s
10:44<Magotari-#uml->>Right. So just stick to qemu then.
10:44<Magotari-#uml->>Erm, sorry.
10:44<Magotari-#uml->>I mean Xen.
10:44<Magotari-#uml->>Hell, you could just use Xen exclusively, if performance is critical.
10:45<jcpiza-#uml->>or KVM
10:46|-|Infinito [argos@200-140-154-162.gnace701.dsl.brasiltelecom.net.br] has quit [Quit: Quitte]
10:46<jcpiza-#uml->>but, i never did run JVM inside of virtual machine
10:47<Magotari-#uml->>What does JVM have to do with anything? It is a userspace program.
10:47<Magotari-#uml->>It won't even know it is in a virtual machine.
10:47<jcpiza-#uml->>Magotari, native theads
10:47<Magotari-#uml->>Oh.
10:48<dang-#uml->>KVM, not JVM
10:48<Magotari-#uml->>Well, if you are running in Xen I don't think it will matter. The operating system with hardware extensions does not even know it is virtualized.
10:48<jcpiza-#uml->>how can i measure the performance of Apache Tomcat webserver?
10:48<dang-#uml->>KVM is the kernel virtualization project. It has a version of qemu that works with it.
10:48<Magotari-#uml->>Ditto for KVM. Same technology.
10:49<Magotari-#uml->>I think KVM can also emulate itself. Not sure if they put that in.
10:49<Magotari-#uml->>I know they had plans. I stopped watching a while back, got better (uml) stuff to do. Especially that I don't have a recent cpu.
10:50<dang-#uml->>I stopped watching it because qemu is a royal pain to use.
10:50<dang-#uml->>I just stick to UML/chroot.
10:50<dang-#uml->>(and I didn't care about non-linux OSs)
10:50|-|hfb [~hfb@75.80.37.175] has quit [Quit: Leaving]
10:50<jcpiza-#uml->>a Tomcat (Java JVM) webserver in Xen,Qemu,KVM or VirtualBox and 100 linux clients from VNUML. Is this scenario possible?
10:50<Magotari-#uml->>jcpiza: Yes.
10:51<Magotari-#uml->>If you get VMUML to behave, that is. I don't think highly of that project.
10:51<Magotari-#uml->>qemu can be a paint.
10:51<Magotari-#uml->>*pain.
10:51<Magotari-#uml->>I have a dvorak keyboard layout. Qemu always gives me qwerty. Ugh.
10:51<jcpiza-#uml->>qemu is slower
10:51<jcpiza-#uml->>pcbochs slowest!
10:52<Magotari-#uml->>Really, if performance matters so much, forget anything but os virtualization and Xen.
10:52<Magotari-#uml->>Maybe KVM, but you are already losing performance there.
10:53<jcpiza-#uml->>Magotari, i want to measure those performances in virtualizated environments, not in real environments.
10:56<Magotari-#uml->>Xen makes even the host virtual.
10:58<jcpiza-#uml->>hehehe, web providers provide virtual hostings, not real hostings.
11:39|-|jdike [~jdike@pool-71-248-190-161.bstnma.fios.verizon.net] has joined #uml
11:39<jdike-#uml->>Hi guys
11:40<Magotari-#uml->>Hey, jdike.
11:41<jdike-#uml->>Magotari, about your question about changelogs last night
11:41<jdike-#uml->>they should describe what's being changed and why
11:42<jdike-#uml->>and probably shouldn't say "I tested this for 6 hours, but I'm a moron, so maybe it's still completely wrong"
11:43<Magotari-#uml->>Right. Why not warn the people though? I am a potential danger, and while I have best hopes to improve things...
11:44<jdike-#uml->>changelogs will live on kernel.org forever
11:44<jdike-#uml->>so do you want your warnings to be in the kernel tree for posterity?
11:45<Magotari-#uml->>I don't mind. Truth is one thing I highly value.
11:45<Magotari-#uml->>But I guess it is kinda silly.
11:45<Magotari-#uml->>By now I know C somewhat well...
11:46<jdike-#uml->>there are true things you don't need to say all the time
11:47<jdike-#uml->>also, if you feel strongly about it, put such comments in [] at the beginning of the changelog
11:47<jdike-#uml->>those will tend to be removed before being committed
11:47<Magotari-#uml->>[memory hole]. I will keep this in mind.
11:47<Magotari-#uml->>Thanks for advice.
11:48|-|hfb [~hfb@pool-71-106-219-180.lsanca.dsl-w.verizon.net] has joined #uml
11:48<Magotari-#uml->>Have you seen the file at Documentation/uml/ ? I have the formatting displaying quite strangely in nano and ed.
11:49<Magotari-#uml->>I even wrote a script to clean up the ^H appearing in the text, but there are lots of other formatting matters which beg for attention.
11:51|-|dgraves [~agraves@inet-netcache2-o.oracle.com] has quit [Remote host closed the connection]
11:52<Magotari-#uml->>Did you generate the file from something? I kinda got such an impression.
11:52<Magotari-#uml->>And the text does sound familiar, like I read it before.
11:58<jdike-#uml->>it's generated from the same XML as the web site
11:59<Magotari-#uml->>I have this in my dmesg: line_ioctl: tty0: unknown ioctl: 0x5600
11:59<Magotari-#uml->>This is VT_OPENQRY
11:59<Magotari-#uml->>I got it when trying to run X in uml.
12:00<Magotari-#uml->>I think this is a defconfig kernel. rc3-mm2.
12:00<Magotari-#uml->>I don't think a bug, but struck me as interesting.
12:02<jdike-#uml->>can it be implemented in UML?
12:02|-|dgraves [~agraves@inet-netcache2-o.oracle.com] has joined #uml
12:03|-|dgraves_ [~agraves@inet-netcache2-o.oracle.com] has joined #uml
12:03<jdike-#uml->>eek
12:03<Magotari-#uml->>I have no idea, I have not researched this.
12:03<jdike-#uml->>dgraves is cloning himself
12:03<Magotari-#uml->>Heh.
12:04<dgraves_-#uml->>jdike, its almost working.
12:04<dgraves_-#uml->>:)
12:04|-|dgraves [~agraves@inet-netcache2-o.oracle.com] has left #uml []
12:04|-|dgraves_ changed nick to dgraves
12:04<dgraves-#uml->>i figure i can get more done that way.
12:13<Magotari-#uml->>grep FIXME time. Plus stress testing of the new SKAS3 patch.
12:16<Magotari-#uml->>And crashme.
12:17<Magotari-#uml->>Pretty nice, with all that load CFS keeps the system pretty smooth.
12:23|-|ram [~ram@pool-96-225-204-220.ptldor.fios.verizon.net] has joined #uml
12:23|-|flatronf700B [~flatronf7@ns1.clipsalportal.com] has quit [Ping timeout: 480 seconds]
12:27<Magotari-#uml->>jdike: Found something. A system is still up, but it does not look too well. After running crashme for a while I am getting floating point exceptions when trying to nano something. Same for even 'init 0'.
12:27<jdike-#uml->>hmm
12:27<Magotari-#uml->>Gonna try without skas3
12:27<jdike-#uml->>anything you run gets an FPE?
12:27<Magotari-#uml->>Not everything. ed worked fine.
12:28<Magotari-#uml->>More things run than not.
12:28<Magotari-#uml->>man man worked... ls worked...
12:28<Magotari-#uml->>Hmm...
12:29<Magotari-#uml->>I have strace output for the machine.
12:29<Magotari-#uml->>I put strace inside my uml, and the last thing before it gives me a sigfpe for init 0 is set_thread_area
12:30<jdike-#uml->>I'd expect a SIGSEGV if something went wrong with set_thread_area
12:30<Magotari-#uml->>For nano this is not so, the last thing is an ioctl, "SNDCTL_TMR_TIMEBASE or TCGETS" as strace told me.
12:30<jdike-#uml->>not SIGFPE
12:31<Magotari-#uml->>Also, in either case the operation is successful.
12:31<Magotari-#uml->>I got a zero from the ioctl and the set_thread_area
12:32<Magotari-#uml->>Wish I had copy-paste here. :/
12:34<Magotari-#uml->>And I finished the machine off with gdbing init 0. sysrq t gives me a huge stack.
12:34<Magotari-#uml->>For init.
12:55<dgraves-#uml->>Magotari, rafb.net
12:56<Magotari-#uml->>dgraves: Yeah, that is not the problem.
12:56<Magotari-#uml->>The terminal I use does not allow copying.
12:57<dgraves-#uml->>Magotari, wow.
12:57<Magotari-#uml->>xterm or eterm both suck that way, no easy way out. eterm at least tries, but I had uml on an xterm.
12:57<dgraves-#uml->>what terminal.
12:57<dgraves-#uml->>er?
12:57<dgraves-#uml->>xterm has copy paste.
12:57<Magotari-#uml->>SHIT.
12:57<dgraves-#uml->>just highlight, middle mouse button.
12:57<Magotari-#uml->>Oh. I tried that. Did not work.
12:58<Magotari-#uml->>Or was that for pasting...
12:58<Magotari-#uml->>Not like it matters now, with the vm off, but next time...
12:58<dgraves-#uml->>Magotari, highlight automatically copies to the X clipboard.
12:58<dgraves-#uml->>middle mouse pastes to another application.
13:05[~]jdike #uml bisects his quilt series looking for where gdb broke#uml-> bisects his quilt series looking for where gdb broke
13:07[~]dgraves #uml wonders how to do that. ;)#uml-> wonders how to do that. ;)
13:07<jdike-#uml->>hehe
13:08<Magotari-#uml->>jdike: Everything works fine on my system now.
13:08<Magotari-#uml->>It was the crashme run, then stuff started happening.
13:08<Magotari-#uml->>I did not write down the argements I invoked it with. Stupid.
13:44|-|krau [~cktakahas@200.184.118.132] has quit [Quit: Varei!!!]
13:44|-|krau [~cktakahas@200.184.118.132] has joined #uml
13:58|-|aindilis [~aindilis@75.146.96.197] has quit [Remote host closed the connection]
14:46|-|da-x [karrde@bzq-79-179-107-6.red.bezeqint.net] has joined #uml
14:56|-|tyler29 [~tyler@ARennes-257-1-160-169.w86-214.abo.wanadoo.fr] has joined #uml
15:33|-|Micksa_ [~mslade@203.55.194.112] has quit [Ping timeout: 480 seconds]
15:40|-|cl4sh [~cl4sh@qik.ds.pg.gda.pl] has joined #uml
15:44|-|ferret_0567 [~travis@cpe-72-191-26-86.satx.res.rr.com] has quit [Quit: leaving]
15:44|-|ferret_0567 [~travis@72.191.26.86] has joined #uml
15:48<jdike-#uml->>6 off to akpm
15:50|-|Magotari [~karol@chello089076073248.chello.pl] has quit [Remote host closed the connection]
15:54|-|Micksa [~mslade@203.55.194.112] has joined #uml
15:56|-|da-x [karrde@bzq-79-179-107-6.red.bezeqint.net] has quit [Remote host closed the connection]
15:56|-|Magotari [~karol@chello089076073248.chello.pl] has joined #uml
16:59|-|tyler29 [~tyler@ARennes-257-1-160-169.w86-214.abo.wanadoo.fr] has quit [Ping timeout: 480 seconds]
17:27|-|dang [~dang@aa-redwall.nexthop.com] has quit [Quit: Leaving.]
17:45<dgraves-#uml->>jdike, still running! :)
17:59|-|cl4sh [~cl4sh@qik.ds.pg.gda.pl] has quit [Quit: leaving]
18:12|-|hfb [~hfb@pool-71-106-219-180.lsanca.dsl-w.verizon.net] has quit [Quit: Leaving]
18:26|-|newbie [~jeff@ita4fw1.itasoftware.com] has joined #uml
18:33|-|dang [~dang@nemesis.fprintf.net] has joined #uml
19:19|-|Infinito [argos@200-140-154-162.gnace701.dsl.brasiltelecom.net.br] has joined #uml
19:38<Magotari-#uml->>Playing around with mconsole. Adding a 'kill <pid>' thingy for fun and training. Won't be useful to anyone, but it is something simple I can do.
19:41<Magotari-#uml->>Something to keep to myself, anyway.
19:55|-|IntuitiveNipple [~TJ@alexandros.tjworld.net] has quit [Quit: The only intuitive interface is the nipple; everything else is learned]
20:55<Magotari-#uml->>Wow.
20:55<Magotari-#uml->>It actually nearly works. I killed my sshd.
20:56<Magotari-#uml->>Yup. It works.
20:57<Magotari-#uml->>Would be nice to specify a signal in general though, -9 is not needed all the time...
20:57<Magotari-#uml->>Erm, 9 I mean.
21:15<Magotari-#uml->>3:13am. Oh dear.
21:30|-|namit [~test_acco@83-70-39-95.b-ras1.prp.dublin.eircom.net] has quit [Ping timeout: 480 seconds]
21:57|-|Infinito [argos@200-140-154-162.gnace701.dsl.brasiltelecom.net.br] has quit [Quit: Quitte]
22:55|-|fiddo [~fiddo@220-253-140-169.WA.netspace.net.au] has joined #uml
22:55<fiddo-#uml->>hi there is this the channel about the unified modelling language UML ?
23:11<caker-#uml->>fiddo: read /topic
23:14<fiddo-#uml->>ah
23:14|-|fiddo [~fiddo@220-253-140-169.WA.netspace.net.au] has left #uml []
23:36|-|balbir [~balbir@122.167.179.91] has quit [Ping timeout: 480 seconds]
23:58|-|VS_ChanLog [~stats@ns.theshore.net] has left #uml [Rotating Logs]
23:59|-|VS_ChanLog [~stats@ns.theshore.net] has joined #uml
---Logclosed Thu Dec 13 00:00:27 2007