Back to Home / #uml / 2007 / 06 / Prev Day | Next Day
#uml IRC Logs for 2007-06-19

---Logopened Tue Jun 19 00:00:14 2007
00:30|-|ram [] has joined #uml
02:53|-|motp[hi] [~motp@] has quit [Quit: Leaving]
03:10|-|cmantito [] has quit [Quit: I aer quitted.]
06:38|-|aroscha [] has quit [Quit: aroscha]
07:33|-|kokoko1 [~Slacker@] has joined #uml
07:42|-|aroscha [] has joined #uml
07:42|-|aroscha_ [] has joined #uml
07:42|-|aroscha [] has quit [Read error: Connection reset by peer]
07:54|-|aroscha_ [] has quit [Ping timeout: 480 seconds]
07:59|-|aroscha [~aroscha@] has joined #uml
08:12|-|aroscha [~aroscha@] has quit [Ping timeout: 480 seconds]
08:16|-|aroscha [] has joined #uml
09:52|-|jdike [] has joined #uml
09:52<jdike>Hi guys
09:54|-|silug [] has joined #uml
09:55<jdike>kokoko1, so
09:55<jdike>that UML still behaving itself?
09:56<kokoko1>up 5 days, 20:47
09:56<kokoko1>It was apache + fedora problem not uml :)
09:57<kokoko1>hmm duno but i also increased MaxClients which was set to 100 before to 10000
09:57<jdike>any idea what the real problem was?
09:57<kokoko1>may be it helps
09:57<kokoko1>jdike, I did few things and now i am not sure which things fix the problem , uml kernel, removed duplicated packages and changed MaxClients in httpd.conf
09:59<kokoko1>i am sure it wasn't uml kernel fault, coz it was going unresponsive even in latest kernel
09:59<kokoko1>surely something to do with apache :)
10:00<jdike>well, that could just mean there's an unfixed bug
10:24|-|Nem^ [] has joined #uml
10:25|-|hfb [] has joined #uml
11:01|-|the_hydra [~a_mulyadi@] has joined #uml
11:01<the_hydra>kokoko1: you there?
11:01<the_hydra>jdike: hi jeff
12:01|-|aroscha [] has quit [Quit: aroscha]
12:08|-|ram [] has quit [Ping timeout: 480 seconds]
12:08|-|mttbtt-3 [] has joined #uml
12:18|-|Nem^ [] has quit [Quit: Send me a postcard, you know you want to >>]
12:19|-|ram [] has joined #uml
12:29|-|hfb [] has left #uml [Leaving]
12:43|-|the_hydra [~a_mulyadi@] has quit []
13:00|-|mttbtt-3 [] has quit [Quit: Leaving]
13:24|-|dgraves [] has joined #uml
13:24<dgraves>I LIVEEEE!!!!!!!!!!
13:24<dgraves>morning all.
13:24<dgraves>jdike: is there a current feature list for uml?
13:35<jdike>not as such
13:35<jdike>there's the documentation
13:35<jdike>why do you need one?
13:49<kokoko1>went to sleep i thinks :)
13:50|-|_horst [] has joined #uml
14:02|-|kos_tom [] has joined #uml
14:13|-|_horst [] has quit [Remote host closed the connection]
14:14<dgraves>jdike: we were interested in what we were missing out on in the later versions of uml. :)
14:15<dgraves>(ie, i'm trying to build a business case for the upgrade)
14:15<dgraves>and i thought a feature list would be a neat thing to have.
14:17<jdike>what are you running now?
14:18<jdike>a good amount of performance, for sure
14:18<jdike>dunno how much that matters to you
14:25<dgraves>very much.
14:25<dgraves>jdike, but a lot of our stuff is disk I\O, so i don't think it would help much.
14:26<jdike>plus a whole lot of bug fixes
14:27<jdike>are you actually I/O bound on the host?
14:27<dgraves>for most of our tests, yes.
14:27<jdike>FWIW, a kernel build inside UML is CPU-bound
14:27<dgraves>we're doing heavy filesystem tests.
14:28<jdike>and it's maybe 30-40% faster than it used to be
14:28<dgraves>yeah, i know. we do things like write tons of data randomly, read it, etc. its more than a kernel build, but probably not quite io bound.
14:28<dgraves>hmm.... that's a benefit right there. :)
14:28<jdike>and I just discovered that I was kmallocing kernel stacks
14:28<dgraves>i was just thinking, you know, a list of things uml does, like smp, etc, like we used to have on the website for features.
14:29<jdike>changing that to get_free_pages gave me what looks like another few %
14:29<dgraves>not bad.
14:30<dgraves>every few % adds up.
14:30<jdike>currently around 70% of native
14:30<jdike>working on 75%
14:30<dgraves>wow, not bad.
14:31<dgraves>that's up there with xen and vmware.
14:31<dgraves>you got a few more with KVM stuff, didn't you?
14:31[~]dgraves loves performance. :)
14:31<jdike>haven't run it under kvm yet
14:31<dgraves>did .21 get smp for skas?
14:31<jdike>I'd hope for more than a few
14:31<dgraves>just curious. :)
14:31<jdike>there are lots of SMP cleanups, but the actual SMP stuff, no
14:35|-|Electric1lf [] has joined #uml
14:38|-|ElectricElf [] has quit [Ping timeout: 480 seconds]
15:33|-|richardw [] has joined #uml
16:39|-|kos_tom [] has quit [Remote host closed the connection]
16:46|-|richardw [] has quit [Quit: Leaving]
16:51|-|krau [~cktakahas@] has quit [Ping timeout: 480 seconds]
17:02|-|krau [~cktakahas@] has joined #uml
17:14|-|_horst [] has joined #uml
17:24|-|albertito [] has quit [Quit: q]
17:29|-|albertito [] has joined #uml
17:33|-|krau [~cktakahas@] has quit [Quit: Varei!!!]
17:33|-|krau [~cktakahas@] has joined #uml
17:34|-|albertito [] has quit [Remote host closed the connection]
17:39|-|albertito [] has joined #uml
17:50|-|_horst [] has quit [Remote host closed the connection]
17:55|-|alb [] has joined #uml
17:56|-|albertito [] has quit [Quit: Reconnecting]
20:39|-|jdike [] has quit [Quit: Leaving]
20:54|-|ram [] has quit [Ping timeout: 480 seconds]
22:59|-|VS_ChanLog [] has left #uml [Rotating Logs]
22:59|-|VS_ChanLog [] has joined #uml
---Logclosed Wed Jun 20 00:00:47 2007