TheGrebs.com:IRC Logs:#xen:2006:06:19

---	Log	opened Mon Jun 19 00:00:18 2006
01:19	\|-\|	mastermind [~mastermin@mastermind.kaltenbrunner.cc] has quit [Quit: Client exiting]
01:22	\|-\|	gerrit [~gerrit@67.160.146.170] has joined #xen
02:10	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has quit [Quit: foo]
02:18	\|-\|	henning__ changed nick to lazyb0y
02:29	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
02:35	\|-\|	mejlholm [~mejlholm@scenic76.cs.aau.dk] has joined #xen
02:50	\|-\|	pvanhoof [~pvanhoof@62.58.98.194] has joined #xen
02:57	\|-\|	Tv [~tv@i1.inoi.fi] has joined #xen
03:09	\|-\|	sdog [~sdog@206.99-201-80.adsl.skynet.be] has joined #xen
03:25	\|-\|	womble [~mpalmer@220.233.135.106] has quit [Quit: What's behind the round window...]
03:37	\|-\|	athomas [~athomas@host86-136-84-147.range86-136.btcentralplus.com] has joined #xen
03:57	\|-\|	mejlholm [~mejlholm@scenic76.cs.aau.dk] has quit [Quit: Leaving]
03:59	\|-\|	visik7 [~visi@155.185.28.97] has joined #xen
04:05	\|-\|	tessier [~treed@75.5.99.178] has joined #xen
04:05	\|-\|	athomas [~athomas@host86-136-84-147.range86-136.btcentralplus.com] has quit [Quit: Leaving]
04:09	\|-\|	athomas [~athomas@host86-136-84-147.range86-136.btcentralplus.com] has joined #xen
04:10	\|-\|	horms [~horms@210.128.90.14] has quit [Quit: Leaving]
04:30	\|-\|	tessier [~treed@75.5.99.178] has quit [Quit: Leaving]
05:09	\|-\|	cableman [~cableman@62.107.91.69] has joined #xen
05:10	\|-\|	lipeng [~pli1@pdfwpr01.pd.intel.com] has quit [Quit: Download Gaim: http://gaim.sourceforge.net/]
05:41	\|-\|	tessier [~treed@75.5.99.178] has joined #xen
05:54	\|-\|	pvanhoof [~pvanhoof@62.58.98.194] has quit [Remote host closed the connection]
05:59	\|-\|	dezent [dezent@unixgeek.biz] has joined #xen
06:01	<dezent>	hello.. im planning 2 xenservers.. i would like to use 3.0.2, i would prefer to use debian but it seems theres no stable packages available at this time.. should i go with fedora instead ?
06:03	<koki>	dezent: try http://www.backports.org/
06:03	<koki>	there just cannot be stable packages in debian
06:04	<koki>	policy renders this impossible to add packages to sarge
06:05	<dezent>	yeah i know.. but sortof stable backports
06:05	<dezent>	just like backports.org
06:05	<dezent>	ive used backports for long time now
06:05	<dezent>	but i didnt know they had xen
06:05	<dezent>	hm is there a packaged 2.6.16-xen kernel aswell ?
06:05	<dezent>	cant find it
06:10	<koki>	dezent: called linux-image now
06:10	<dezent>	aah
06:10	<dezent>	thanx alot
06:10	<dezent>	=)
06:12	\|-\|	horms [~horms@p7209-ipbf22marunouchi.tokyo.ocn.ne.jp] has joined #xen
06:19	\|-\|	visik7 [~visi@155.185.28.97] has quit [Remote host closed the connection]
06:38	\|-\|	parabelboi [~parabelbo@blueice4n1.de.ibm.com] has joined #xen
06:58	<netman1>	dezent: Just for encourgement: running Debian Sarge + backports.org xen, very stable. Just 2 tips: remove hotplug before installing xen and fix udev through /etc/apt/preferences to 0.091-1bpo1 and all should run fine. Have fun!
06:58	<dezent>	netman1: great !
06:59	<dezent>	im installing the first server right now
06:59	<koki>	same for me, works like a charm, even the migration.
07:00	<dezent>	sounds great.. havnt been using redhat since linuxppc1999 on my old Amiga
07:00	<dezent>	hm
07:03	\|-\|	horms [~horms@p7209-ipbf22marunouchi.tokyo.ocn.ne.jp] has quit [Quit: Leaving]
07:14	<dezent>	netman1: should the stuff in /etc/apt/preferences look something like this ?
07:14	<dezent>	Package: udev
07:14	<dezent>	Pin: release a=sarge-backports
07:14	<dezent>	Pin-Priority: 999
07:17	<netman1>	Yes, ..
07:17	<dezent>	ok thnx
07:17	<netman1>	but add a ", v=0.091-1bpo1" to the Pin line.
07:19	<netman1>	Package: udev
07:19	<netman1>	Pin: release a=sarge-backports v=0.091-1bpo1
07:19	<netman1>	Pin-Priority: 999
07:28	<dezent>	huge thanx
07:30	\|-\|	bernarde [~bernarde@143.106.24.24] has joined #xen
07:31	\|-\|	womble [~mpalmer@150.101.204.102] has joined #xen
07:37	\|-\|	womble [~mpalmer@150.101.204.102] has quit [Quit: What's behind the round window...]
07:53	\|-\|	johnlev [~johnlev@nwkea-socks-2.sun.com] has joined #xen
07:59	\|-\|	pvanhoof [~pvanhoof@d5152D086.access.telenet.be] has joined #xen
08:04	\|-\|	SLot [~SLot___@gustavo.gruposim.com.br] has quit [Quit: Fudeu geral hehehehe]
08:30	\|-\|	fabien [~fabien@193.54.76.165] has joined #xen
08:30	<fabien>	hy
08:31	<fabien>	my station reboot when I have a kernel panic, so i can't see the error message, is there a solution to avoid the reboot ?
08:33	<sdague>	append "noreboot" to the xen line in menu.lst
08:34	<fabien>	on the "kernel ..." line or "module ..." line plz ?
08:36	<fabien>	.. and google said "on the kernel line", thank you & google
08:38	<dezent>	is there any howto for installing the packages from backports.org ?
08:38	<dezent>	think ill need one
08:40	\|-\|	unriel changed nick to riel
08:41	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has joined #xen
08:41	\|-\|	SLot [~SLot___@gustavo.gruposim.com.br] has joined #xen
08:46	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has quit [Quit: Client exiting]
08:52	\|-\|	jimix [~jimix@ip139.152.45.216.suscom.net] has joined #xen
08:53	<fabien>	It work fine sdague, but I can't see the whole call trace, is there a tip to log that (sorry, i am a newby concerning kernel hacking)
09:03	\|-\|	hollisb [~hollisb@192.35.232.241] has joined #xen
09:04	<fabien>	bye
09:04	\|-\|	fabien [~fabien@193.54.76.165] has quit [Quit: Quitte]
09:07	\|-\|	parabelboi [~parabelbo@blueice4n1.de.ibm.com] has quit [Ping timeout: 480 seconds]
09:12	\|-\|	rharper [~rharper@pixpat.austin.ibm.com] has joined #xen
09:18	\|-\|	parabelboi [~parabelbo@195.212.29.187] has joined #xen
09:22	\|-\|	deshantm [~deshantm@bi-02pt1.bluebird.ibm.com] has joined #xen
09:26	\|-\|	parabelboi [~parabelbo@195.212.29.187] has quit [Read error: Connection reset by peer]
09:30	\|-\|	Tv [~tv@i1.inoi.fi] has quit [Quit: foo]
09:37	\|-\|	Astinus [~astinus@astinus.usercloak.oftc.net] has quit [Ping timeout: 480 seconds]
09:38	\|-\|	schultmc [~schultmc@216.37.46.162] has joined #xen
09:42	\|-\|	parabelboi [~parabelbo@blueice2n1.de.ibm.com] has joined #xen
09:57	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has joined #xen
10:04	\|-\|	bernarde [~bernarde@143.106.24.24] has quit [Remote host closed the connection]
10:15	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has joined #xen
10:16	\|-\|	jasonmc [jason@spoon.netsoc.tcd.ie] has joined #xen
10:17	\|-\|	ns [~nivedita@67.171.181.157] has quit [Quit: The voices in my head are telling me to...]
10:28	\|-\|	pvanhoof [~pvanhoof@d5152D086.access.telenet.be] has quit [Remote host closed the connection]
10:28	\|-\|	stekloff [~stekloff@bi01p1.co.us.ibm.com] has joined #xen
10:29	\|-\|	jdmason [~jonmason@192.35.232.241] has joined #xen
10:31	\|-\|	ns [~niv@bi01p1.co.us.ibm.com] has joined #xen
10:32	\|-\|	mastermind [~mastermin@mastermind.kaltenbrunner.cc] has joined #xen
10:36	<jasonmc>	how do i find out what version of xen particular source code is
10:36	<jasonmc>	(i'm looking for a hypervisor&kernel to go with 3.0.2-8 userspace)
10:37	<hollisb>	jasonmc: did you use mercurial to get the source?
10:37	<jasonmc>	no, i used the website (testing)
10:37	\|-\|	jerone [~jerone@pixpat.austin.ibm.com] has joined #xen
10:37	<hollisb>	I don't understand. you're fine with reinstalling kernel and hypervisor, but you don't want to touch the userspace tools?
10:37	<jasonmc>	we already have built an rpm for the tools...
10:38	<hollisb>	so you can rpm --rebuild then, right? :)
10:38	<jasonmc>	yea
10:38	<jasonmc>	but how do i find out what version particluar source is?
10:39	<jasonmc>	what i might be doing is taking the xen rpm from FC5 and building a kernel&hypervisor for it
10:39	<jasonmc>	it's just easier that way
10:40	<jasonmc>	what would be great, however, is if there were rpms available for x86_64 RHEL4 !
10:40	<hollisb>	you're using a snapshot tarball?
10:40	<jasonmc>	yes
10:41	<hollisb>	then I don't think you have any guarantees about compatibility. for example, an interface could have changed but a full release wasn't made
10:41	<hollisb>	in which case the makefile would still say it's 3.0.2 or whatever
10:42	<jasonmc>	oh, i see. thanks anyway
10:43	<hollisb>	the compatibility guarantees are only about guest domains
10:43	<hollisb>	the maintainers have explicitly stated that you must be running matched tools and hypervisor
10:43	<jasonmc>	ok. well, that's the problem i ran in to anyway...
10:44	<jasonmc>	xend not being able to run
10:44	<hollisb>	right. so always install xend and the hypervisor at the same time from the same source
10:45	<jasonmc>	got ya
10:45	<jasonmc>	are there even source rpms available anywhere?
10:45	<hollisb>	from the distros
10:45	\|-\|	jerone [~jerone@pixpat.austin.ibm.com] has quit [Quit: Leaving]
10:46	<jasonmc>	it's a shame xensource don't distribute the ones they use to build the RHEL 4 i386 rpms
10:46	\|-\|	jerone [~jerone@pixpat.austin.ibm.com] has joined #xen
10:47	<hollisb>	jasonmc: why don't you ask for them?
10:47	<jasonmc>	good idea!
10:51	\|-\|	rpg [~rpg@192.35.232.241] has joined #xen
10:55	<athena>	jasonmc: let us know what they say pls :)
10:56	<jasonmc>	yea, i've just dropped them an email
10:57	\|-\|	plars [~plars@pixpat.austin.ibm.com] has joined #xen
11:00	\|-\|	stekloff [~stekloff@bi01p1.co.us.ibm.com] has quit [Remote host closed the connection]
11:03	\|-\|	bernarde [~bernarde@201.44.214.48] has joined #xen
11:07	\|-\|	stekloff [~stekloff@bi01p1.co.us.ibm.com] has joined #xen
11:07	<jasonmc>	this may be a silly question but, is it safe to run recent kernels form kernel.org without updating all the userspace stuff?
11:07	<jasonmc>	(i don't mean xen, just in general)
11:07	<mastermind>	jasonmc: depends
11:08	<jasonmc>	it would make the system unstable, right...
11:08	<mastermind>	things like LVM or udev might break
11:09	<jasonmc>	yea, i expect so.
11:09	<mastermind>	jasonmc: what is you base OS ?
11:09	<jasonmc>	is it possible then, to patch an older kernel version with xen
11:09	<jasonmc>	scientific linux 4 (basically RHEL4)
11:37	\|-\|	athomas [~athomas@host86-136-84-147.range86-136.btcentralplus.com] has quit [Remote host closed the connection]
11:38	\|-\|	muli [~muli@87.69.40.180] has joined #xen
11:40	\|-\|	parabelboi [~parabelbo@blueice2n1.de.ibm.com] has quit [Read error: Connection reset by peer]
11:46	\|-\|	Marco [~marco@249-174.surfsnel.dsl.internl.net] has joined #xen
11:46	<Marco>	hi
11:57	\|-\|	tobi [~tobi@zux006-049-116.adsl.green.ch] has joined #xen
12:04	\|-\|	jasonmc [jason@spoon.netsoc.tcd.ie] has left #xen [Leaving work]
12:06	\|-\|	Basic_py [~Basic@warden.real-time.com] has joined #xen
12:15	\|-\|	gerrit [~gerrit@67.160.146.170] has quit [Ping timeout: 480 seconds]
12:40	<movement>	hollisb: there?
12:46	<hollisb>	movement: yup
12:50	<movement>	so, with your 32-bit tools, do you have an API or whatever for getting the word size of xen?
12:50	<hollisb>	no, I pretty much just made everything 64-bit quanities
12:51	<hollisb>	so 32-bit userland tools are using uint64_t
12:51	<movement>	I was asking because the error message you get trying to boot the wrong class of domU is so freaking awful
12:51	<hollisb>	I'm not sure what you mean by "wrong class"
12:52	<movement>	32-bit domu on 64-bit hypervisor
12:52	<movement>	and vice versa. you get: Error: (9, 'Bad file number')
12:52	<hollisb>	:)
12:53	<hollisb>	so that would be xc_linux_load.c? the domU builder?
12:53	<movement>	and ERROR: Kernel not a Xen-compatible Elf image. in xend-debug.log
12:53	<movement>	hollisb: I presume so
12:54	<hollisb>	I should say that my patches didn't extend to the guest/hypervisor interface
12:55	<hollisb>	I couldn't change that and maintain binary compat without an even bigger mess
12:55	<movement>	I'm not sure what you mean? the builder isn't a guest/hypervisor interface
12:56	<movement>	I know that non-matching domU's can't boot
12:56	<hollisb>	well, you're trying to load a 32-bit kernel on a 64-bit hypervisor
12:56	<hollisb>	ok...
12:56	<hollisb>	so it sounds like the error message should be improved, but it should still be an error
12:57	<hollisb>	and this particular area wasn't one that I got into
12:57	<movement>	yes
12:57	<movement>	I was only wondering if you'd needed a similar API in the stuff you'd been doing at some point, that's all
12:57	<hollisb>	ah I see
12:58	<hollisb>	I don't think so. I think fixed size structures are better here
12:58	<movement>	certainly
13:16	\|-\|	gerrit [~gerrit@bi01p1.co.us.ibm.com] has joined #xen
13:27	\|-\|	aliguori [~anthony@70.116.9.243] has quit [Quit: Ex-Chat]
13:27	\|-\|	aliguori [~anthony@cpe-70-116-9-243.austin.res.rr.com] has joined #xen
13:39	\|-\|	pvanhoof [~pvanhoof@d54C0E27E.access.telenet.be] has joined #xen
13:59	\|-\|	johnlev_ [~johnlev@nwkea-socks-1.sun.com] has joined #xen
14:01	\|-\|	aliguori [~anthony@cpe-70-116-9-243.austin.res.rr.com] has quit [Quit: Ex-Chat]
14:01	\|-\|	aliguori [~anthony@cpe-70-116-9-243.austin.res.rr.com] has joined #xen
14:05	\|-\|	_plars [~plars@pixpat.austin.ibm.com] has joined #xen
14:05	\|-\|	johnlev [~johnlev@nwkea-socks-2.sun.com] has quit [Ping timeout: 480 seconds]
14:09	\|-\|	plars [~plars@pixpat.austin.ibm.com] has quit [Ping timeout: 480 seconds]
14:11	\|-\|	_plars [~plars@pixpat.austin.ibm.com] has quit [Remote host closed the connection]
14:13	\|-\|	bernarde [~bernarde@201.44.214.48] has quit [Remote host closed the connection]
14:13	\|-\|	bernarde [~bernarde@201.44.214.48] has joined #xen
14:16	\|-\|	bernarde [~bernarde@201.44.214.48] has quit [Remote host closed the connection]
14:17	\|-\|	bernarde [~bernarde@201.44.214.48] has joined #xen
14:18	\|-\|	SLot [~SLot___@gustavo.gruposim.com.br] has quit [Ping timeout: 480 seconds]
14:49	\|-\|	cdub [~chrisw@cdub.netrep.oftc.net] has quit [Ping timeout: 480 seconds]
14:52	<aliguori>	are grant references mappable in userspace?
14:58	\|-\|	visik7 [~visi@host104-47.pool877.interbusiness.it] has joined #xen
14:58	\|-\|	cdub [~chrisw@cdub.netrep.oftc.net] has joined #xen
15:07	\|-\|	aliguori [~anthony@cpe-70-116-9-243.austin.res.rr.com] has quit [Quit: Ex-Chat]
15:10	\|-\|	aliguori [~anthony@cpe-70-116-9-243.austin.res.rr.com] has joined #xen
15:15	\|-\|	JViz [Anomaly@adsl-065-013-131-023.sip.int.bellsouth.net] has joined #xen
15:23	\|-\|	DV_ [~veillard@veillard.com] has joined #xen
15:24	\|-\|	bernarde [~bernarde@201.44.214.48] has quit [Remote host closed the connection]
15:25	\|-\|	bernarde [~bernarde@201.44.214.48] has joined #xen
15:25	\|-\|	sdog [~sdog@206.99-201-80.adsl.skynet.be] has quit [Ping timeout: 480 seconds]
15:26	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has quit [Quit: Client exiting]
15:35	\|-\|	nixx [devnull@h209-71-198-8.gtconnect.net] has joined #xen
15:42	\|-\|	visik7 [~visi@host104-47.pool877.interbusiness.it] has quit [Remote host closed the connection]
15:45	\|-\|	nixx [devnull@h209-71-198-8.gtconnect.net] has quit [Quit: Client exiting]
15:45	\|-\|	jerone [~jerone@pixpat.austin.ibm.com] has quit [Quit: Leaving]
15:47	\|-\|	visik7 [~visi@host104-47.pool877.interbusiness.it] has joined #xen
15:48	\|-\|	lazyb0y_ [~henning@e176103102.adsl.alicedsl.de] has joined #xen
15:55	\|-\|	lazyb0y [~henning@85.176.110.66] has quit [Ping timeout: 480 seconds]
15:56	\|-\|	visik7 [~visi@host104-47.pool877.interbusiness.it] has quit [Remote host closed the connection]
16:00	\|-\|	bernarde [~bernarde@201.44.214.48] has quit [Remote host closed the connection]
16:00	\|-\|	bernarde [~bernarde@201.44.214.48] has joined #xen
16:03	\|-\|	cableman [~cableman@62.107.91.69] has left #xen [Leaving]
16:03	\|-\|	plars [~plars@pixpat.austin.ibm.com] has joined #xen
16:08	\|-\|	riel changed nick to unriel
16:11	\|-\|	visik7 [~visi@host104-47.pool877.interbusiness.it] has joined #xen
16:21	\|-\|	jerone [~jerone@m015f36d0.tmodns.net] has joined #xen
16:25	\|-\|	hollisb [~hollisb@192.35.232.241] has quit [Quit: leaving]
16:33	\|-\|	Guy- [~korn@chardonnay.math.bme.hu] has joined #xen
16:33	<Guy->	hi
16:34	<Guy->	how would I go about configuring a 32bit xen domU kernel on amd64? make menuconfig ARCH=i386 doesn't show the xen options
16:35	<athena>	i don't know, but i can tell you that just a few hours ago it was pointed out in this very place, that a 32-bit domU and 64-bit hypervisor don't play well :)
16:38	<Guy->	indeed?
16:38	<Guy->	that was what I wanted to try out :)
16:38	<Guy->	in what way don't they play well?
16:39	\|-\|	bernarde [~bernarde@201.44.214.48] has quit [Read error: Connection reset by peer]
16:42	<athena>	the following error (less than lucid) error message was reported when trying to boot: Error: (9, 'Bad file number')
16:42	<athena>	and ERROR: Kernel not a Xen-compatible Elf image. in xend-debug.log
16:42	<Guy->	oh, lovely error message there :)
16:42	<aliguori>	Guy-, it just won't work
16:43	<aliguori>	it's not supported
16:43	[~]	athena scolds the file number, "bad file number!!"
16:43	<Guy->	aliguori: is support planned? I didn't see it in the roadmap, but you never know your luck
16:44	<aliguori>	athena, there's almost no error handling in Xend. and user-friendly messages you see are because somewhere someone proactively checked for a common error. once something actually fails, the error messages get so screwed up in the stack that determining what they actually mean is near impossible
16:44	<aliguori>	Guy-, not as far as I know. it would be pretty difficult to support
16:44	<aliguori>	if you have VT/SVM, you can run a 32 bit guest in HVM mode
16:45	<Guy->	alas, no
16:45	<athena>	aliguori: is that because it's difficult to handle errors or because it just wasn't done?
16:45	<Guy->	but I can still run 32bit apps on a 64bit guest kernel, right?
16:45	<aliguori>	athena, it's just crufty code
16:45	<aliguori>	Guy-, yes
16:45	<athena>	ah, thanks
16:47	<athena>	guy-: see http://lists.xensource.com/archives/html/xen-devel/2005-10/msg00037.html
16:49	<Guy->	thanks
16:52	<Guy->	it's not entirely clear to me yet what the 'backend' and 'frontend' drivers are; am I right in assuming that a domU kernel generally doesn't need backend drivers?
16:53	\|-\|	DV_ [~veillard@veillard.com] has quit [Quit: Leaving]
16:55	<Guy->	also, I'm using kernel-package from Debian unstable to build the xen kernels, and I need to specify --arch amd64 --subarch xen, otherwise it passes ARCH=xen to make, which then fails; is this a bug in make-kpkg, do you think?
16:56	\|-\|	rharper [~rharper@pixpat.austin.ibm.com] has quit [Quit: Leaving]
17:17	\|-\|	jimix [~jimix@ip139.152.45.216.suscom.net] has quit [Quit: jimix]
17:19	\|-\|	pvanhoof [~pvanhoof@d54C0E27E.access.telenet.be] has quit [Remote host closed the connection]
17:26	\|-\|	plars [~plars@pixpat.austin.ibm.com] has quit [Remote host closed the connection]
17:58	\|-\|	jerone [~jerone@m015f36d0.tmodns.net] has quit [Quit: Leaving]
18:16	\|-\|	womble [~mpalmer@106.135.233.220.exetel.com.au] has joined #xen
18:33	\|-\|	Shoragan_ [~shoragan@d072.apm.etc.tu-bs.de] has quit [Quit: Leaving]
18:43	\|-\|	Basic_py [~Basic@warden.real-time.com] has quit [Quit: Leaving]
18:46	\|-\|	MarkW [~MarkW@maw48.kings.cam.ac.uk] has joined #xen
18:46	\|-\|	mode/#xen [+o MarkW] by ChanServ
18:47	<@MarkW>	Lo, all.
18:48	<aliguori>	hey MarkW!
18:49	<@MarkW>	aliguori: Hey
18:49	<@MarkW>	Long time no see
18:49	<aliguori>	yeah
18:49	<aliguori>	how's life treatin' ya?
18:49	\|-\|	deshantm [~deshantm@bi-02pt1.bluebird.ibm.com] has quit [Ping timeout: 480 seconds]
18:49	<@MarkW>	Oh, not too bad :-)
18:49	<aliguori>	you gonna make it out to OLS this year?
18:49	<@MarkW>	Don't think so this time
18:50	<aliguori>	:-/ that's ashame
18:50	<aliguori>	how's xenfs coming?
18:50	<@MarkW>	:-(
18:50	<@MarkW>	Oh, XenFS is noodling along slowly
18:50	<@MarkW>	The VFS API really hurts my head :-)
18:50	<aliguori>	hehe
18:50	<@MarkW>	Wish I was using FUSE. But of course, that would be less fun
18:51	<aliguori>	yeah, I think filesystems are almost life long commitments :-) quite complicated
18:51	<aliguori>	mm, fuse..
18:51	<aliguori>	who needs performance anyway :-)
18:51	<aliguori>	there's also p9fs
18:51	<@MarkW>	Yes
18:51	<aliguori>	i think that's supposed to be more reasonable right?
18:51	<@MarkW>	Which seems interesting but I've not had a chance to go into in detail yet
18:51	<@MarkW>	Might start trying to use it as a template
18:52	<@MarkW>	I'm fiddling around with basic write support for XenFS now.
18:53	<@MarkW>	And I'm starting to think that making live migration work with XenFS is possibly more interesting than I'd originally thought.
18:53	<aliguori>	hrm
18:53	<aliguori>	you would make it transparently networked?
18:53	<@MarkW>	Nope
18:54	\|-\|	mr_flea [~mr_flea@cpe-70-113-197-207.austin.res.rr.com] has joined #xen
18:54	<@MarkW>	The idea would be to leave behind all non-dirty XenFS pages and not migrate them at all
18:54	<@MarkW>	=> lower migration latency
18:54	<@MarkW>	But to get the destination node to pre-read them into memory so that the overheads are minimised.
18:55	<@MarkW>	That way you get to migrate without sending the page cache contents, but without the usual performance overheads associated.
19:00	\|-\|	deshantm [~deshantm@bi-02pt1.bluebird.ibm.com] has joined #xen
19:00	\|-\|	JViz [Anomaly@adsl-065-013-131-023.sip.int.bellsouth.net] has quit [Read error: No route to host]
19:02	<aliguori>	interesting
19:08	\|-\|	JViz [Anomaly@adsl-065-013-131-023.sip.int.bellsouth.net] has joined #xen
19:08	<@MarkW>	I'd hope it means we greatly reduce network traffic / latency for a migration, whilst still retaining performance.
19:13	\|-\|	horms [~horms@vagw.valinux.co.jp] has joined #xen
19:14	\|-\|	ns [~niv@bi01p1.co.us.ibm.com] has quit [Quit: The voices in my head are telling me to...]
19:18	\|-\|	stekloff [~stekloff@bi01p1.co.us.ibm.com] has quit [Quit: Leaving]
19:27	\|-\|	horms [~horms@vagw.valinux.co.jp] has quit [Quit: Leaving]
19:28	\|-\|	horms [~horms@vagw.valinux.co.jp] has joined #xen
19:42	\|-\|	ns [~nivedita@c-67-171-181-157.hsd1.or.comcast.net] has joined #xen
19:45	\|-\|	hollisb [~hollisb@user-0vvde2g.cable.mindspring.com] has joined #xen
19:48	\|-\|	gerrit [~gerrit@bi01p1.co.us.ibm.com] has quit [Ping timeout: 480 seconds]
19:58	\|-\|	hollisb [~hollisb@user-0vvde2g.cable.mindspring.com] has quit [Quit: Lost terminal]
20:30	\|-\|	JViz [Anomaly@adsl-065-013-131-023.sip.int.bellsouth.net] has quit [Ping timeout: 481 seconds]
20:30	\|-\|	JViz [Anomaly@adsl-065-013-131-023.sip.int.bellsouth.net] has joined #xen
20:37	\|-\|	lipeng [~pli1@pdfwpr01.pd.intel.com] has joined #xen
21:00	\|-\|	lipeng [~pli1@pdfwpr01.pd.intel.com] has quit [Quit: Download Gaim: http://gaim.sourceforge.net/]
21:09	<tessier>	horms: You rock
21:12	<horms>	tessier: thanks. morning to you too :)
21:12	<tessier>	:)
21:13	<tessier>	horms: What sort of work are you doing with xen?
21:13	\|-\|	cdub_ [~chrisw@216-99-217-87.dsl.aracnet.com] has joined #xen
21:13	<horms>	most of my xen work relates to either porting kexec from linux, or build system fixes
21:14	<tessier>	Back when I worked at mp3.com I used perdition on the mail server. Now we use linux-ha and lvs at my current shop. I am deploying xen now also. Looking for an excuse to use super sparrow. :)
21:14	<horms>	hehe, super sparrow needs a bit more love than i have time to get it :(
21:15	\|-\|	cdub [~chrisw@cdub.netrep.oftc.net] has quit [Ping timeout: 480 seconds]
21:15	<tessier>	kexec is scary. If you don't have everything set up just right your system goes poof. If you can't afford a reboot I don't see how one could risk a kexec.
21:16	<horms>	its more to do with the speed of the reboot, rather than the riskiness
21:16	<tessier>	And suddenly microkernels sound like a better idea.
21:16	<horms>	also, some hw doesn't reliably reboot
21:16	<tessier>	So you can crash even faster?
21:17	<horms>	but the real reason for my efforts is the kdump portion of kexec, which hopwfully allows you to take a core when the hypervisor or dom0 crash
21:17	<tessier>	Cool. I am surprised that hasn't been possible before now.
21:17	<horms>	in practice i find kexec quite useful for development purposes, but i'm not sure how useful it would be in other areas
21:18	<tessier>	Linus has always been in favor of printk's and catching stuff on the console it seems.
21:18	<surriel>	as long as we don't put kexec into Xen, I'm ok with it
21:18	<surriel>	but apparently :(((((
21:18	<horms>	a lot of coporations here in japan want to do forensics in their kernel, which is why its a kind of hot topic here. developers tend to be more intersted in writing correct core
21:18	<surriel>	whoever promised to put kexec into Xen must not have heard about virtualization
21:18	<horms>	code
21:19	<surriel>	just put the damn crashdump kernel into a separate domain
21:19	<horms>	surriel: what if the hypervisor crashes?
21:19	<surriel>	then it can't schedule the crashdump kernel!
21:20	<tessier>	I've had some weird problems tracking down the source of disk activity lately. Are there any tools which will show you which domain is causing lots of disk activity? And any way to characterize the disk activity? Such as reads or seeks etc?
21:20	<horms>	usually when a crash occurs, there is some life left inorder to triger a crashdump
21:31	<tessier>	And hope it wan't your disk driver that crashed so you can write the dump out
21:32	<horms>	well, obviously there are problems with where to put the dump
21:48	\|-\|	Netsplit charon.oftc.net <-> venus.oftc.net quits: Hunger
21:49	\|-\|	Netsplit charon.oftc.net <-> quasar.oftc.net quits: sdague, lazyb0y_, lilo2, womble, jforbes, @MarkW, netman1, minemaz, horms, ns
21:50	\|-\|	Netsplit over, joins: ns, horms, MarkW, womble, lazyb0y_, minemaz
21:50	\|-\|	Netsplit over, joins: lilo2, sdague, netman1
21:50	\|-\|	Netsplit over, joins: jforbes
21:50	\|-\|	NetHack #xen [+o MarkW] by [quasar.oftc.net]
21:50	\|-\|	Netsplit over, joins: Hunger
22:02	\|-\|	Basic_py [~Basic@warden.real-time.com] has joined #xen
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has left #xen [Rotating Logs]
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
23:12	\|-\|	bartman_ [~bart@CPE0014bfaf982f-CM0011ae8a8ba6.cpe.net.cable.rogers.com] has joined #xen
23:27	\|-\|	tobi [~tobi@zux006-049-116.adsl.green.ch] has quit [Ping timeout: 480 seconds]
23:49	\|-\|	gerrit [~gerrit@c-67-160-146-170.hsd1.or.comcast.net] has joined #xen
---	Log	closed Tue Jun 20 00:00:46 2006