Back to Home / #uml / 2007 / 01 / Prev Day | Next Day
#uml IRC Logs for 2007-01-24

---Logopened Wed Jan 24 00:00:49 2007
00:46|-|less_ [] has quit [Ping timeout: 480 seconds]
00:51|-|less_ [] has joined #uml
02:12|-|peterz [] has joined #uml
02:16<peterz>does anybody know if jeff made his uml smp patches public?
02:20<caker>I haven't seen them anywhere
02:20<peterz>ok, I'll just poke jeff when he gets online again
02:35|-|motp [~motp@] has joined #uml
02:38|-|ram [] has quit [Ping timeout: 480 seconds]
03:19|-|fbenites [~chatzilla@] has joined #uml
03:25|-|jeroen_ [] has joined #uml
03:25<jeroen_>has anyone ever been crazy enough to try UML on Xen ?
03:25<peterz>s/UML on// no
03:26<jeroen_>well perhaps it's just a different problem: no matter what kernel or hostsystem I try, I always get a host mount error...
03:26|-|Urgleflogue [] has quit [Read error: Connection reset by peer]
03:27<jeroen_>I mean root fs mounting error
03:28<fbenites>jeroen_: do you have hostfs?
03:28<fbenites>on the guest system
03:28<jeroen_>well, I tried both hostfs and booting from an image
03:29<jeroen_>actually last time I had a different error
03:29<jeroen_>Kernel panic - not syncing: start_userspace : expected SIGSTOP, got status = 2943
03:29<jeroen_>after the root was mounted read-only
03:32<fbenites>maybe is your image wrong?
03:32<fbenites>but i never heard of this problem
03:34|-|cmantito [] has quit [Quit: I aer quitted.]
03:38|-|less_ [] has quit [Ping timeout: 480 seconds]
03:39<jeroen_>I downloaded the image from
03:40<jeroen_>is it still true that the host kernel has to be or lower?
03:41|-|cmantito [] has joined #uml
03:41[~]peterz runs -git host and guest
03:41|-|Urgleflogue [] has joined #uml
03:42<jeroen_>and I saw some mention of problems with glibc 2.4 and up?
03:42<peterz>fixed since
03:46|-|less_ [] has joined #uml
03:50|-|tyler [~tyler@] has joined #uml
04:20|-|Urgleflogue [] has quit [Ping timeout: 480 seconds]
04:25|-|Urgleflogue [] has joined #uml
04:45<fbenites>does somebody have experience with using usb drivers on the uml kernel?
04:50|-|tyler [~tyler@] has quit [Ping timeout: 480 seconds]
05:35|-|tyler [~tyler@] has joined #uml
05:43|-|tyler [~tyler@] has quit [Ping timeout: 480 seconds]
06:10|-|tyler [~tyler@] has joined #uml
08:07|-|motp [~motp@] has quit [Quit: Leaving]
08:47|-|tyler [~tyler@] has quit [Ping timeout: 480 seconds]
08:59|-|fbenites [~chatzilla@] has quit [Remote host closed the connection]
09:00|-|tyler [~tyler@] has joined #uml
09:10|-|fbenites [~chatzilla@] has joined #uml
09:20|-|jdike [] has joined #uml
09:20<jdike>Hi guys
09:20<peterz>hey Jeff
09:21<peterz>jdike: do you have those smp patches in a public place?
09:21<peterz>I'd like to have a play with it
09:22<jdike>I can put that tree on the patches page, I guess
09:22<jdike>it's not the cleanest of trees
09:23<peterz>awell, as long as it sorta does what its supposed to do
09:23<jdike>it does
09:24<peterz>see, no problems ;-)
09:24<peterz>worst that can happen is that it breaks, in which case I'll just keep the pieces
09:25<peterz>anyway, gotta pick up some groceries, bbiab
09:26|-|less_ [] has quit [Quit: Leaving]
10:53<peterz>jdike: put the patches up somewhere?
10:54<jdike>nope, not yet
10:55<peterz>ok, then its not my incompetence keeping me from finding them ;-)
10:55<jeroen_>jdike: uml on xen seems to be a nono
10:56<jdike>there was a recent bug fix which might help that
10:56<jeroen_>are you talking to peterz or to me ?
10:57<peterz>I gather you, since a bugfix for incompetence would be awesome ;-)
10:58<jeroen_>I tried the latest kernel though, but I couldn't even build a um kernel under a software xen
10:58<jeroen_>and all precompiled kernels had problems with mounting the root filesystem
10:59<jeroen_>I finally figured out how to do hardware xen now though, that's also a solution
11:01<jeroen_>I could try to test again in a couple of days when you have a kernel source for me
11:03<jeroen_>anyway, I gotta go, been a long day, see you later
11:03|-|jeroen_ [] has quit [Quit: jeroen_]
11:26|-|nite613 [] has joined #uml
11:26<nite613>Hi there
11:27<nite613>I have a question about tun/tap. Can I connect more than one UML to a given TAP device, or do I need to run a uml_switch instance and connect THAT?
11:27<jdike>the second
11:28<nite613>Figured :( Okay, Running 64 different uml_switch instances it is, then :(
11:29<jdike>how many UMLs are you running?
11:29<jdike>or is it two, each with 64 eth devices?
11:36<nite613>2, each with 64 devices
11:36<nite613>It's a network security lab where each student gets their own device and some tools tied to it
11:36|-|HuK0B [~HuK0B@] has joined #uml
11:39|-|albertito [] has quit [Read error: Connection reset by peer]
11:40<HuK0B>jdike is it possible uml to have problem with time intervals
11:40<jdike>what's the real question?
11:40|-|albertito [] has joined #uml
11:40<jdike>it shouldn't
11:41<HuK0B>I mean in one of my mail server I set time interval to execute queue 100mins but it execute it for 25mins
11:41<HuK0B>this is not really my problem I mean I host a machine to one friend and he told me about this problem
11:42<jdike>it's supposed to run the queue every 100 min?
11:42<HuK0B>there was other problem with ntp timeout that was set was 60 seconds and I had problems with that it worked ok when I set it to 6000
11:42<jdike>and it really runs every 25 mintes
11:42<jdike>UML version?
11:42|-|albertito [] has quit [Read error: Connection reset by peer]
11:43<HuK0B>don't know he report me that is atound 25mins
11:43<HuK0B>uml kernel
11:43<jdike>can you find a simpler case which has a timing problem?
11:43<jdike>like sleep n not sleeping for n seconds?
11:44<HuK0B>hmm I will go to that mashine and will try sleep(60) function
11:44<HuK0B>if this ok?
11:45<jdike>yeah, time sleep 60
11:45<jdike>should say something really close to 60
11:47<HuK0B>hmm no problem at my umls
11:47<HuK0B>ftp:/ftp/ftp/ time sleep 60
11:47<HuK0B>real 1m0.075s
11:47<HuK0B>user 0m0.010s
11:47<HuK0B>sys 0m0.000s
11:47<HuK0B>may be he have some ok
11:48|-|albertito [] has joined #uml
11:49<jdike>there was a problem a while back where sleep n was really sleep n - 1
11:49<jdike>which meant that sleep 1 finished really fast
11:49<jdike>but that's been fixed for a while
11:50<HuK0B>I didn't understand you
11:50<HuK0B>but nvm
11:52|-|tyler [~tyler@] has quit [Ping timeout: 480 seconds]
11:53|-|tyler [~tyler@] has joined #uml
11:58|-|the_hydra [~mulyadi@] has joined #uml
12:30|-|fbenites [~chatzilla@] has quit [Remote host closed the connection]
12:52|-|ram [] has joined #uml
13:06|-|HuK0B [~HuK0B@] has quit [Read error: Connection reset by peer]
13:11|-|tyler [~tyler@] has quit [Ping timeout: 480 seconds]
13:12<caker>jdike: a little late, but: g'day mate
13:13<caker>jdike: have a good trip/time?
13:16<dgraves>the_hydra: hey, try using a piece of paper or something useful instead of a ball in that HZ example.
13:16<the_hydra>kindly explain, please?
13:20|-|tyler [~tyler@] has joined #uml
13:23<the_hydra>dgraves: kindly explain pls?
13:27<dgraves>oh, instead of using a ball that ...
13:27<dgraves>you throw around.
13:27<dgraves>you use something like i don't know, chopsticks that you pass around.
13:27<dgraves>a ball doesn't leave much to do with it.
13:27<dgraves>but Hz settings affect how much work you can get done with something.
13:28<the_hydra>trying to get your idea...
13:28<the_hydra>maybe it is like paper, get passed between busy writers
13:29<the_hydra>the shorter the HZ, the less a writer can write on to this paper
13:29<the_hydra>sorry, the bigger
13:29<the_hydra>so, if it's so big, the writer just keep passing it without actually write something
13:30<the_hydra>is that what you are trying to say?
13:30<dgraves>the_hydra: sure. but if its bigger, the writer just keeps writing, not passing it.
13:30<the_hydra>Corect Me If I am Wrong
13:30<dgraves>the_hydra: yeah, that's exactly what i'm saying. just a better thought than a ball.
13:31|-|da-x_ [] has quit [Ping timeout: 480 seconds]
13:31<the_hydra>i try to adopt it
13:31<the_hydra>but actually HZ is not directly related to rescheduling
13:32<the_hydra>it's more like an alarm buzzed directly to your ear
13:32<the_hydra>the more it buzzes, the more you get "shocked"
13:32<the_hydra>hence illustrating the interrupt, interrupting your concentration
13:40<dgraves>fair enough.
13:40<dgraves>it does affect more than just rescheduling, i guess.
13:41<the_hydra>btw, how is your day over there allan?
13:41<dgraves>its going.
13:41<dgraves>almost done projects. :)
13:41<dgraves>which is good.
13:42<dgraves>cause i had to tell boss why one would be late. :-(
13:42<the_hydra>i see
13:42<dgraves>yeah. don't like that. :(
13:43<dgraves>how about you?
13:43<the_hydra>not so fun, my mind stucks while the deadline is approaching
13:43<the_hydra>and this saturday, i turn into 28 years old
13:44<peterz>ooh, congratz
13:45<the_hydra>yes yes thanks peter
13:58<dgraves>the_hydra: congrats!
13:58<dgraves>you're almost as old as me. :)
13:58<the_hydra>thanks a lot !
13:58<dgraves>i'm 29. :)
13:58<peterz>I'll still be 28 saturday
13:58<peterz>I'll turn 29 during OLS
13:59<dgraves>very nice.
14:00<the_hydra>ok, so soon enough, I will join club 30s
14:00<the_hydra>dgraves: gee, I thought you were around 40!
14:02[~]dgraves wonders if that's good or bad.
14:03<peterz>just tell yourself you make a mature impression ;-)
14:03<the_hydra>according to Einstein, it's all about relativity
14:03<the_hydra>so I think 40 is good :)
14:08<dgraves>the_hydra: intersting. most people say I'm immature.
14:08<dgraves>wonder if its like in person verse text.
14:08<the_hydra>yes, especially after reading your jokes in IRC :D
14:08<peterz>either that, or like the_hydra suggested its relative, which says something about the_hydra ;-)
14:09<nite613>jdike: I Set up the uml_switch and my UML seems to be connecting to it (uml_switch process says "New connection" when I start the UML and shows some MAC addresses), but I can't pass packets
14:10<nite613>I get "send_sock sending to fd 4 Resource temporarily unavailable" from the uml_switch process when I try to send a ping
14:28|-|karrde_ [] has joined #uml
14:33|-|the_hydra [~mulyadi@] has quit [Remote host closed the connection]
14:48|-|hfb [] has joined #uml
14:59|-|tyler [~tyler@] has quit [Ping timeout: 480 seconds]
15:08<jdike>nite613, what are you trying to ping?
15:13<nite613>I got it all sorted out. Apparently it was working okay and then a script later in my boot process ws runing ifconfig on non-existent interfaces and hanging it up
15:13<nite613>That's a problem I can avoid, if not explain
15:19<jdike>is the hanging it up a boot script problem or a UML problem?
16:10<nite613>.... an hour later
16:11<nite613>it seems like it could be a UML problem. The boot script works if the interfaces exist, but if they don't it hangs while running ifconfig on them
16:11<nite613>No matter, it's a silly thing to do
16:11<jdike>what happens if you run it by hand in a shell?
16:12<jdike>does the UML hang, or just the ifconfig?
16:16|-|hfb [] has left #uml [Leaving]
17:19|-|jdike [] has quit [Quit: Leaving]
17:36|-|ram [] has quit [Ping timeout: 480 seconds]
17:45|-|ram [] has joined #uml
18:19|-|albertito [] has quit [Read error: Connection reset by peer]
18:21|-|albertito [] has joined #uml
21:22|-|ram [] has quit [Ping timeout: 480 seconds]
22:35|-|Nem^1 [] has joined #uml
22:43|-|Nem^ [] has quit [Ping timeout: 480 seconds]
22:43|-|Nem^1 changed nick to Nem^
22:59|-|VS_ChanLog [] has left #uml [Rotating Logs]
22:59|-|VS_ChanLog [] has joined #uml
23:44|-|albertito [] has quit [Ping timeout: 480 seconds]
---Logclosed Thu Jan 25 00:00:38 2007