TheGrebs.com:IRC Logs:#xen:2006:01:28

---	Log	opened Sat Jan 28 00:00:09 2006
00:15	\|-\|	f [~andi@p5489BB3A.dip0.t-ipconnect.de] has joined #xen
00:21	\|-\|	xterminus [nobody@00104bc8bd59.click-network.com] has quit [Ping timeout: 480 seconds]
00:59	\|-\|	gerrit [~gerrit@c-24-22-19-162.hsd1.or.comcast.net] has quit [Ping timeout: 480 seconds]
01:06	\|-\|	gerrit [~gerrit@c-24-22-19-162.hsd1.or.comcast.net] has joined #xen
01:20	\|-\|	gerrit [~gerrit@c-24-22-19-162.hsd1.or.comcast.net] has quit [Ping timeout: 480 seconds]
01:32	\|-\|	gerrit [~gerrit@c-24-22-19-162.hsd1.or.comcast.net] has joined #xen
01:54	\|-\|	f [~andi@p5489BB3A.dip0.t-ipconnect.de] has quit [Quit: Hey! Where'd my controlling terminal go?]
01:57	\|-\|	jayesh [~jayesh@c-24-118-178-182.hsd1.mn.comcast.net] has joined #xen
01:58	\|-\|	wenchien [~wenchien@221-169-69-23.adsl.static.seed.net.tw] has quit [Ping timeout: 480 seconds]
01:58	\|-\|	gerrit [~gerrit@c-24-22-19-162.hsd1.or.comcast.net] has quit [Ping timeout: 480 seconds]
02:23	\|-\|	Shaun2222 [~Shaun@ip68-5-63-223.oc.oc.cox.net] has joined #xen
02:31	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
02:32	\|-\|	pvanhoof [~pvanhoof@d54C0FBBD.access.telenet.be] has joined #xen
02:58	\|-\|	jayesh [~jayesh@c-24-118-178-182.hsd1.mn.comcast.net] has quit [Remote host closed the connection]
03:12	\|-\|	xterminus [nobody@00104bc8bd59.click-network.com] has joined #xen
03:28	\|-\|	Basic_py [~Basic@warden.real-time.com] has joined #xen
04:12	\|-\|	Netsplit helium.oftc.net <-> oxygen.oftc.net quits: weasel, andy
04:14	\|-\|	Netsplit over, joins: weasel, andy
05:35	\|-\|	tarawa1943 [~tarawa194@c-67-169-198-176.hsd1.or.comcast.net] has joined #xen
05:38	\|-\|	jesse_ [~wenchien@221-169-69-23.adsl.static.seed.net.tw] has joined #xen
05:38	\|-\|	jesse_ changed nick to wenchien
05:48	\|-\|	DV [~veillard@veillard.com] has quit [Ping timeout: 480 seconds]
05:51	\|-\|	DV [~veillard@veillard.com] has joined #xen
06:22	\|-\|	weasel [weasel@weasel.noc.oftc.net] has quit [Quit: reboot]
08:12	\|-\|	tarawa1943 [~tarawa194@c-67-169-198-176.hsd1.or.comcast.net] has quit [Quit: Leaving]
08:18	\|-\|	johnlev [~johnlev@nwkea-socks-1.sun.com] has quit [Ping timeout: 480 seconds]
09:12	\|-\|	monrad_ [~mikkel@213083190131.sonofon.dk] has joined #xen
09:13	\|-\|	monrad [~mikkel@213083190131.sonofon.dk] has quit [Read error: Connection reset by peer]
09:56	\|-\|	muli [~muli@87.69.40.180.cable.012.net.il] has quit [Quit: My damn controlling terminal disappeared!]
12:01	<m0dY>	Hey all,, i wanna make a small POLL here ! i'm workin on a community site which will be including a well tested working images - domU's - for both i386 & x86_64 machines/kernels for free to be downloaded & will be inclduing some tutorial and helpfull stuff about xen for beginnners !! So will this be a good idea to complete on !?
12:06	<m0dY>	not interesting !?
12:06	<grifferz>	personally I am capable of making my own images when I need them..
12:06	<grifferz>	a site of documentation on how to do so would be more useful to me
12:07	<grifferz>	hmm that didn't make too much sense
12:07	<grifferz>	I WANT to make my own images whenever possible, and usually can. if I were having trouble I'd rather have some docs on how to do it rather than a site that lets me download the image itself
12:07	<m0dY>	well,, sometimes you face problems installing redhat like systems on debians and vice versa
12:08	<grifferz>	in which case I'd rather google and find your site telling me how to do it, not your site saying "here's one I made earlier" :)
12:09	<m0dY>	well, there'll be both .
12:09	<m0dY>	the reason for images is for people wanna have a good start in a couple of minutes
12:10	<m0dY>	so is there any idea i can get it on the site ?
12:10	<m0dY>	something helpfull for the xen community
12:42	<aliguori>	m0dY, this already exists
12:42	<aliguori>	let me dig up the url
12:43	<aliguori>	m0dY, www.xen-get.org
12:43	<aliguori>	you can contribute images to that site if you'd like
12:43	<aliguori>	that would certainly be appreciated
12:50	\|-\|	monrad_ [~mikkel@213083190131.sonofon.dk] has quit [Read error: Connection reset by peer]
12:59	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has quit [Ping timeout: 480 seconds]
13:07	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
13:10	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has joined #xen
13:14	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has quit [Quit: mdday]
14:47	\|-\|	gerrit [~gerrit@c-24-22-19-162.hsd1.or.comcast.net] has joined #xen
14:56	\|-\|	monrad [~mikkel@213083190131.sonofon.dk] has joined #xen
15:17	\|-\|	gerrit [~gerrit@c-24-22-19-162.hsd1.or.comcast.net] has quit [Ping timeout: 480 seconds]
15:18	\|-\|	yorkedork [~thom@adsl-221-53-102.rmo.bellsouth.net] has joined #xen
15:19	\|-\|	yorkedork [~thom@adsl-221-53-102.rmo.bellsouth.net] has quit [Quit: ]
15:44	\|-\|	gerrit [~gerrit@c-24-22-19-162.hsd1.or.comcast.net] has joined #xen
15:52	\|-\|	Sir_Ahzz [~ahzz@c-24-0-241-237.hsd1.tx.comcast.net] has quit [Remote host closed the connection]
15:53	\|-\|	Sir_Ahzz [~ahzz@c-24-0-241-237.hsd1.tx.comcast.net] has joined #xen
15:53	\|-\|	mode/#xen [+o Sir_Ahzz] by ChanServ
16:16	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has quit [Quit: Leaving]
17:20	\|-\|	Brian_L [~brian@c-69-137-239-43.hsd1.md.comcast.net] has joined #xen
17:28	\|-\|	Brian_L [~brian@c-69-137-239-43.hsd1.md.comcast.net] has quit [Quit: Leaving]
18:02	\|-\|	hikenboot [~hikenboot@c-24-218-84-234.hsd1.ma.comcast.net] has joined #xen
18:03	<hikenboot>	hello all .....whats new?
18:03	<hikenboot>	bsd is going to support xen 3.0 in its upcoming 6.1 release and on 64bit platform for 6.2 release
19:34	<Shaun>	does xenstore store cpu_time?
19:35	<aliguori>	Shaun, no
19:38	<Shaun>	aliguori: i'm having a hell of a time trying to get cpu % usage
19:39	<aliguori>	Shaun, give me a minute...
19:39	<Shaun>	i keep getting values higher than 1.00
19:39	<aliguori>	oh
19:39	<Shaun>	i think my problem is that i'm not using --long
19:39	<aliguori>	what are you doing?
19:39	<aliguori>	no
19:39	<aliguori>	it shouldn't be
19:39	<aliguori>	you don't want to parse --long :-)
19:39	<aliguori>	it's a PITA
19:39	<aliguori>	how are you sampling?
19:39	<aliguori>	are you doing:
19:40	<aliguori>	xm list... sleep 1, xm list ....
19:40	<Shaun>	well --long has a longer cpu value...
19:40	<Shaun>	http://pastebin.com/528193
19:40	<aliguori>	highoy
19:40	<aliguori>	oy
19:40	<Shaun>	aliguori: i'm not doing a sleep 1, no need to with microseconds..
19:40	<aliguori>	even
19:41	<aliguori>	i don't read perl well
19:41	<Shaun>	:)
19:41	<aliguori>	you should sleep in between
19:41	<aliguori>	and get the interval before doing the second one
19:42	<aliguori>	that way you're sampling times consistently
19:42	<Shaun>	aliguori: i've moved it around a bunch
19:42	<Shaun>	i've tryed before, after, in the middle, etc :)
19:42	<Shaun>	i'll put it back in their
19:42	<aliguori>	http://pastebin.com/528195
19:42	<aliguori>	i dunno perl
19:42	<Shaun>	my values are looking like this now... 1.3xxx 1.4xxxx
19:42	<aliguori>	so s/sleep(1)/ with whatever sleeping for about a second is
19:43	<Shaun>	the guest is doing a perl -e 'while(1) {}' to consume 100%
19:43	<aliguori>	does tv_interval return a floating point?
19:43	<aliguori>	yeah
19:43	[~]	aliguori breaks out his perl book
19:43	<Shaun>	aliguori: ya thats how i have it right now
19:43	<Shaun>	just changed it to that
19:44	<aliguori>	yeah, xm list doesn't return very high precision
19:44	<aliguori>	you may want to hack the xm code...
19:44	<aliguori>	i can give you a pointer
19:44	<aliguori>	one sec
19:44	<Shaun>	aliguori: thats why i was looking --long
19:45	<Shaun>	xm list --long, (cpu_time 92707.3685138) compared to xm list 92724.8
19:46	<aliguori>	Shaun, if you look at tools/python/xen/xm/main.py: xm_brief_list
19:46	<aliguori>	let me show you what to change it to
19:46	\|-\|	m0dY [m0dY@62.139.160.212] has quit [Ping timeout: 480 seconds]
19:47	<aliguori>	http://pastebin.com/528199
19:47	<aliguori>	remove the 7.1 for cpu_time
19:47	<aliguori>	and you'll get the same thing as you'd get with the s-expression output
19:48	<Shaun>	ya, i suppose i can do that, i really hate having to patch the source though, would rather keep a std tree..
19:49	<Shaun>	aliguori: are you a developer for xen?
19:49	<Shaun>	programmer or whatever..
19:51	<aliguori>	yessir
19:51	<aliguori>	there's another mode
19:51	<aliguori>	one sec
19:52	<Shaun>	you guys need any servers? like sponsers?
19:53	<aliguori>	you'd have to ask one of the cambridge guys...
19:53	<aliguori>	i work for ibm
19:53	<Shaun>	aliguori: we sponder a server for CentOS, we host their site,forums,email,etc....
19:53	<Shaun>	sponser*
19:53	<Shaun>	aliguori: i see
19:53	<aliguori>	you can send a note to Ian Pratt..
19:54	<aliguori>	having a mirror for releases would probably be a really good thing
19:54	<aliguori>	since there were bandwidth concerns for the 3.0.0 release
19:54	<Shaun>	ok, well that change to the source didnt really make much of a diffrence.
19:54	<aliguori>	i'm sure that would be very appreciated :-)
19:54	<aliguori>	changing it to %f?
19:54	<aliguori>	it should.
19:54	<aliguori>	did you make install?
19:55	<Shaun>	aliguori: no, it worked... but it didnt change my calc
19:55	<Shaun>	still getting like 1.3's
19:55	<aliguori>	oh
19:55	<aliguori>	:-)
19:55	<Shaun>	i wonder wtf is going on here...
19:55	<aliguori>	are you on an smp system?
19:55	<Shaun>	this is the problem i had when trying to calculate cpu with jiffies
19:55	<Shaun>	aliguori: no not at the moment.
19:56	<Shaun>	aliguori: actually it should be, it's a 3.0 E so it has HT...
19:56	<Shaun>	proc is only using 1 cpu
19:56	<Shaun>	wonder if i forgot to build xen with SMP support.
19:56	<aliguori>	oh
19:56	\|-\|	m0dY [m0dY@62.139.160.212] has joined #xen
19:56	<aliguori>	yes
19:56	<aliguori>	does xm list show the domains running on different cpus?
19:57	<Shaun>	i remembered PAE support though :)
19:57	<Shaun>	na, they're all running on VCPU 1
19:57	<Shaun>	only hae 2 guests and domain0 running though
19:58	<aliguori>	vcpu 1?
19:58	<aliguori>	that's how many vcpus they have
19:58	<aliguori>	take a look at xm vcpu-list
19:58	<Shaun>	oh lol
19:59	<Shaun>	it shows domain-0 twice (vcpu 0 and 1)... also domain-0 is on cpu0 and the other are one 1
20:00	<Shaun>	http://pastebin.com/528215
20:01	<aliguori>	yup
20:01	<aliguori>	domain 0 is dedicated a hyperthread
20:01	<aliguori>	which improves IO performance significantly
20:01	<Shaun>	thats fine
20:01	<Shaun>	i want it to do that
20:01	<aliguori>	sure
20:01	<Shaun>	in the end this will be running on dual xeon's with HT
20:01	<Shaun>	so 4 procs..
20:01	<Shaun>	possibly dual cores..
20:01	<Shaun>	down the road
20:02	<aliguori>	:-)
20:02	<aliguori>	we just finished putting together a 32-way dual core xeon system...
20:02	<Shaun>	my understanding with dual-core is that they'll show up as 8 procs, havnt tryed them yet..
20:03	<aliguori>	yup
20:04	<Shaun>	thats a mean machine
20:05	<Shaun>	how is xentop getting this percentage..
20:05	<aliguori>	exactly as I described...
20:05	<aliguori>	except making C hypercalls
20:05	<Shaun>	it's probably not forking off xm to get the stats though
20:05	<Shaun>	ya..
20:05	<aliguori>	yeah
20:06	<Shaun>	i wonder if doing that is just lagging and the diffrece between time and the actuall results is varying too much
20:08	<aliguori>	give me a few minutes, i'll see if i can take care of this for you...
20:08	<aliguori>	i've got some old code lying around
20:08	<Shaun>	ok
20:14	<aliguori>	ok, let me reboot and test
20:14	\|-\|	aliguori [~anthony@cpe-70-116-13-229.austin.res.rr.com] has quit [Quit: Leaving]
20:29	\|-\|	aliguori [~anthony@cpe-70-116-13-229.austin.res.rr.com] has joined #xen
20:29	<Shaun>	hey..
20:29	<Shaun>	though you might have ran off :)
20:29	<aliguori>	http://www.codemonkey.ws/vm-list.c
20:29	<aliguori>	compile it with
20:30	<aliguori>	gcc -o vm-list vm-list.c -lxenctrl
20:30	<aliguori>	run it with:
20:30	<aliguori>	./vm-list -f "%D %U" -n 5 -s 1 -q
20:30	<aliguori>	that'll spit out 5 samples at 1 second intervals of all the domain ids and percent cpu usages
20:31	<aliguori>	my wireless doesn't work under xen :-)
20:31	<aliguori>	-f is the format parameter.. it supports a lot of stuff, vm-list --help will show you everything
20:32	<aliguori>	vm-list is from vm-tools. vm-list, at least, still works although most other things in vm-tools don't anymore
20:39	<Shaun>	aliguori: sorry, one sec, had a phone call
20:39	<Shaun>	testing now.
20:40	<Shaun>	oh i see
20:40	<Shaun>	vm-tools is from like xen2 or somthing?
20:41	<aliguori>	yeah
20:41	<aliguori>	or early versions of 3.0
20:45	<Shaun>	aliguori: this was written by you?
20:45	<aliguori>	yup
20:48	<Shaun>	can i pass a username to it
20:48	<aliguori>	username?
20:48	<Shaun>	lol sorry, dom name
20:48	<aliguori>	oh
20:48	<aliguori>	no
20:48	<aliguori>	it's pre-store
20:49	<aliguori>	./vm-list $(xm domid DomainName)
20:49	<Shaun>	with my setup i set the dom names to the username
20:49	<aliguori>	would work
20:54	<Shaun>	this should work awsome, thanks man!
20:55	<aliguori>	np
20:59	\|-\|	weasel [weasel@asteria.debian.or.at] has joined #xen
21:08	<Shaun>	wonder if murble ended up with a patch for the pts stuff
21:16	<Shaun>	aliguori: is their a way to see IO OPS per guest?
21:16	\|-\|	zwane [~zwane@S0106001217d33fa6.vc.shawcable.net] has quit [Remote host closed the connection]
21:16	<aliguori>	Shaun, nope
21:26	<Shaun>	aliguori: any idea if xen is built with ebtables support enabled?
21:27	<Shaun>	aliguori: i'm assuming that xen has the same problem that uml has with bridge interfaces that traffic goes to all interfaces and can be sniffed... i corrected this before with ebtables filters that filtered based off MAC and ip
21:31	<Shaun>	doesnt look like it.
21:40	\|-\|	zwane [~zwane@S0106001217d33fa6.vc.shawcable.net] has joined #xen
21:44	<aliguori>	ebtables?
21:45	<Shaun>	aliguori: http://ebtables.sourceforge.net/
21:45	<aliguori>	yeah, i dunno
21:45	<aliguori>	this is the first i've heard of it
21:45	<Shaun>	aliguori: i'm correct that xen has the same problem right?
21:46	<aliguori>	yeah
21:46	<Shaun>	who controls the configs for xen0
21:46	<aliguori>	?
21:46	<Shaun>	they may want to add ebtables as a module
21:46	<aliguori>	oh
21:46	<aliguori>	just submit a patch
21:46	<aliguori>	to xen-devel
21:46	<aliguori>	it'll almost certainly get checked in...
21:47	<Shaun>	ok, i'm running thru a make ARCH=xen menuconfig right now..
21:47	<aliguori>	Shaun, the default config lives in linux-2.6-sparse/arch/asm-xen/configs/.. or something like that
21:48	<Shaun>	hmm, i'm not finding it!
21:49	<Shaun>	n/m
21:49	<Shaun>	i foudn it
21:53	<Shaun>	bahh, now i'm getting this... http://pastebin.com/528284
21:55	<aliguori>	looks like you messed up your config...
21:55	<aliguori>	heading out, bbl
21:55	\|-\|	aliguori [~anthony@cpe-70-116-13-229.austin.res.rr.com] has quit [Quit: Leaving]
21:55	<Shaun>	all i need was a make ARCH=xen menuconfig and added ebtables support
21:59	\|-\|	hikenboot [~hikenboot@c-24-218-84-234.hsd1.ma.comcast.net] has quit [Remote host closed the connection]
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has left #xen [Rotating Logs]
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
23:16	\|-\|	weasel [weasel@weasel.noc.oftc.net] has quit [Read error: Operation timed out]
23:17	\|-\|	weasel [weasel@asteria.debian.or.at] has joined #xen
23:23	\|-\|	Basic_py [~Basic@warden.real-time.com] has quit [Quit: Leaving]
23:52	\|-\|	_David_ [~david@CPE-203-45-91-168.nsw.bigpond.net.au] has joined #xen
---	Log	closed Sun Jan 29 00:00:56 2006