TheGrebs.com:IRC Logs:#xen:2006:02:16

---	Log	opened Thu Feb 16 00:00:26 2006
00:18	\|-\|	rusty [~rusty@bh02i525f01.au.ibm.com] has quit [Ping timeout: 480 seconds]
01:33	\|-\|	cods [~cods@tuxee.net] has quit [Quit: plouf]
01:45	\|-\|	tessier [~treed@rrcs-67-53-110-66.west.biz.rr.com] has joined #xen
01:49	\|-\|	azoca [~jonan@torrent.callisia.com] has quit [Ping timeout: 480 seconds]
01:50	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has quit [Quit: Client exiting]
01:58	\|-\|	azoca [~jonan@torrent.callisia.com] has joined #xen
02:18	\|-\|	cods [~cods@tuxee.net] has joined #xen
02:20	\|-\|	sdog [~sdog@62.58.98.210] has joined #xen
02:34	\|-\|	azoca [~jonan@torrent.callisia.com] has quit [Quit: leaving]
02:55	\|-\|	cods [~cods@tuxee.net] has quit [Quit: replouf]
03:16	\|-\|	cods [~cods@tuxee.net] has joined #xen
03:18	\|-\|	jimix [~jimix@23.94.233.220.exetel.com.au] has joined #xen
03:22	\|-\|	jimix [~jimix@23.94.233.220.exetel.com.au] has quit [Quit: ]
03:31	\|-\|	Shaun [~ndci@ip68-5-63-223.oc.oc.cox.net] has quit [Quit: ]
03:36	\|-\|	cods [~cods@tuxee.net] has quit [Ping timeout: 480 seconds]
03:44	\|-\|	athomas [~athomas@host86-136-81-69.range86-136.btcentralplus.com] has joined #xen
03:48	\|-\|	m0dY [m0dY@62.139.160.211] has joined #xen
03:57	<m0dY>	hello all,, is there any notes about upgrading from xen 3.0.0 to 3.0.1 ????
04:08	\|-\|	Shaun2 [Shaun@tina.ndcservers.net] has quit [Read error: Connection reset by peer]
04:19	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
04:37	\|-\|	pvanhoof [~pvanhoof@mailhost.newtec.be] has joined #xen
04:40	\|-\|	AlexMBas [~Alexandre@200.103.241.137] has quit [Quit: Fui embora]
04:45	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has quit [Quit: mdday]
04:48	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
05:06	\|-\|	pdx6_ [~ballew@screen.sublinear.net] has joined #xen
05:08	\|-\|	pdx6 [~ballew@screen.sublinear.net] has quit [Read error: Connection reset by peer]
05:08	\|-\|	minemaz [~mine@softbank220055148023.bbtec.net] has quit [Read error: Connection reset by peer]
06:03	\|-\|	musashi [~tjohnson@huygens.utulsa.edu] has quit [Quit: musashi]
06:09	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has quit [Ping timeout: 480 seconds]
06:17	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has joined #xen
06:32	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
07:09	\|-\|	LinuxMan [~LinuxMan@ip-114-163.adsl.cheapnet.it] has joined #xen
07:09	<LinuxMan>	hi guys, i installed freebsd 7.0 on my xen 3.0 but i need to resize loop filesystem 500mbyte to 2000mbyte i don't know how i can do it, who can help pme?
07:17	\|-\|	LinuxMan [~LinuxMan@ip-114-163.adsl.cheapnet.it] has quit [Quit: ]
07:24	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has quit [Ping timeout: 480 seconds]
07:32	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has joined #xen
07:44	\|-\|	rharper [~rharper@ip70-185-18-43.ma.dl.cox.net] has joined #xen
07:57	\|-\|	jonmason [~jonmason@cpe-67-9-183-161.austin.res.rr.com] has joined #xen
08:03	\|-\|	jonmason [~jonmason@cpe-67-9-183-161.austin.res.rr.com] has quit [Remote host closed the connection]
08:07	\|-\|	ogi_ [~ogi@62.44.101.19] has quit [Ping timeout: 480 seconds]
08:10	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has quit [Quit: mdday]
08:11	\|-\|	ogi [~ogi@felix.openfmi.net] has joined #xen
08:24	\|-\|	rharper [~rharper@ip70-185-18-43.ma.dl.cox.net] has quit [Ping timeout: 480 seconds]
08:32	\|-\|	schultmc [~schultmc@zealot.progeny.com] has quit [Ping timeout: 480 seconds]
08:33	\|-\|	mdday [~mdday@cpe-066-057-025-204.nc.res.rr.com] has joined #xen
08:48	\|-\|	anball [~anball@bi01p1.nc.us.ibm.com] has joined #xen
09:00	\|-\|	mdday_ [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
09:01	\|-\|	mdday [~mdday@cpe-066-057-025-204.nc.res.rr.com] has quit [Ping timeout: 480 seconds]
09:08	\|-\|	dansmith [~dan@bi01p1.co.us.ibm.com] has joined #xen
09:14	\|-\|	gaston_ [~gaston@silberzahn.net] has quit [Remote host closed the connection]
09:15	\|-\|	gaston [~gaston@82.227.13.91] has joined #xen
09:26	\|-\|	schultmc [~schultmc@zealot.progeny.com] has joined #xen
09:27	\|-\|	johnlev [~johnlev@nwkea-socks-1.sun.com] has quit [Ping timeout: 480 seconds]
09:36	\|-\|	jerone [~jerone@pixpat.austin.ibm.com] has joined #xen
09:46	\|-\|	aliguori [~anthony@cpe-70-116-13-229.austin.res.rr.com] has quit [Quit: Ex-Chat]
09:46	\|-\|	rharper [~rharper@pixpat.austin.ibm.com] has joined #xen
09:53	\|-\|	ns [~niv@bi01p1.co.us.ibm.com] has joined #xen
10:06	\|-\|	cableman [~cableman@3E6B5E92.rev.stofanet.dk] has joined #xen
10:11	\|-\|	johnlev [~johnlev@nwkea-socks-1.sun.com] has joined #xen
10:14	\|-\|	aliguori [~anthony@pixpat.austin.ibm.com] has joined #xen
10:16	\|-\|	bernarde [~bernarde@32.104.18.240] has joined #xen
10:16	\|-\|	rharper [~rharper@pixpat.austin.ibm.com] has quit [Quit: Leaving]
10:20	\|-\|	rharper [~rharper@pixpat.austin.ibm.com] has joined #xen
10:22	\|-\|	rpg [~rpg@pixpat.austin.ibm.com] has quit [Ping timeout: 480 seconds]
10:22	\|-\|	bernarde [~bernarde@32.104.18.240] has quit [Remote host closed the connection]
10:30	\|-\|	Basic_py [~Basic@warden.real-time.com] has quit [Quit: Leaving]
10:30	\|-\|	sdog [~sdog@62.58.98.210] has left #xen []
10:35	\|-\|	rharper [~rharper@pixpat.austin.ibm.com] has quit [Quit: Leaving]
10:37	\|-\|	pvanhoof [~pvanhoof@mailhost.newtec.be] has quit [Remote host closed the connection]
10:38	\|-\|	mcr [~mcr@CPE0006b123a026-CM0011aea1b6fc.cpe.net.cable.rogers.com] has joined #xen
10:38	\|-\|	rharper [~rharper@pixpat.austin.ibm.com] has joined #xen
10:38	<mcr>	can I have a network card on a Xen0 that is not configured on the Xen0? i.e. is just bridged to a XenU?
10:38	<mcr>	can I have a network interface on a Xen0 that is vlan tagged, and the bridge a single VLAN to a XenU?
10:40	<mcr>	er, not a single VLAN, but rather, a single lan on a XenU.
10:42	<mcr>	can I have network interfaces on a Xen0 which are entirely virtual? i.e. have no physical hardware at all, but are bridged to various XenU? Has anyone considered building a non-ethernet type interface like this? (i.e. able to mmap() 64k blocks between XenU rather than copy)? Or perhaps is ethernet, but is marked as jumbo ethernet frame capable.
10:43	\|-\|	mdday_ [~mdday@cpe-024-163-120-222.nc.res.rr.com] has quit [Read error: Connection reset by peer]
10:45	<tonfa>	mcr: i don't have enough experience with xen, but it looks like the bridged interface does exactly that
10:47	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
10:50	\|-\|	bernarde [~bernarde@32.104.18.240] has joined #xen
10:51	<mcr>	tonfa, I can bridge the hosts' network to the XenU. It's a straight ethernet bridge. The host has veth0 and vifX stuff.
10:52	<mcr>	I guess that xenbr0 is a bridge of peth0 and many vif's. so I guess I can just not configure eth0 into that bridge.
10:52	<mcr>	or, it would be eth1.
10:53	<mcr>	basically, I want the Xen0 on one network, (inward facing), with the XenU on another network (outward facing)
10:53	\|-\|	rpg [~rpg@pixpat.austin.ibm.com] has joined #xen
10:53	<mcr>	mostly, this is for test configurations, but it is critical to the test that broadcasts from one XenU (on network A) are not seen by a XenU (on network B)
11:01	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has joined #xen
11:07	\|-\|	rpg [~rpg@pixpat.austin.ibm.com] has quit [Ping timeout: 480 seconds]
11:08	\|-\|	pdx6_ changed nick to pdx6
11:10	\|-\|	m0dY [m0dY@62.139.160.211] has quit [Ping timeout: 480 seconds]
11:11	\|-\|	aliguori [~anthony@pixpat.austin.ibm.com] has quit [Quit: Ex-Chat]
11:34	\|-\|	Basic_py [~Basic@66.173.4.234] has joined #xen
11:49	\|-\|	athomas [~athomas@host86-136-81-69.range86-136.btcentralplus.com] has quit [Quit: Leaving]
11:53	\|-\|	Basic_java [~Basic@66.173.4.234] has joined #xen
11:57	\|-\|	Basic_py [~Basic@66.173.4.234] has quit [Ping timeout: 480 seconds]
12:16	\|-\|	aliguori [~anthony@pixpat.austin.ibm.com] has joined #xen
12:29	\|-\|	pvanhoof [~pvanhoof@d54C0FBBD.access.telenet.be] has joined #xen
12:29	\|-\|	schultmc [~schultmc@zealot.progeny.com] has quit [Quit: Client exiting]
12:31	\|-\|	rpg [~rpg@pixpat.austin.ibm.com] has joined #xen
12:36	\|-\|	ax [~andrew@implode.fuckdom.net] has left #xen []
12:38	\|-\|	mef [~mef@targe.CS.Princeton.EDU] has joined #xen
12:50	\|-\|	cableman [~cableman@3E6B5E92.rev.stofanet.dk] has quit [Remote host closed the connection]
12:51	\|-\|	aliguori [~anthony@pixpat.austin.ibm.com] has quit [Quit: Ex-Chat]
12:54	\|-\|	aliguori [~anthony@pixpat.austin.ibm.com] has joined #xen
13:08	\|-\|	schultmc [~schultmc@zealot.progeny.com] has joined #xen
13:19	\|-\|	aliguori [~anthony@pixpat.austin.ibm.com] has quit [Quit: Ex-Chat]
13:21	\|-\|	aliguori [~anthony@pixpat.austin.ibm.com] has joined #xen
13:23	\|-\|	aliguori [~anthony@pixpat.austin.ibm.com] has quit [Quit: ]
13:31	\|-\|	aliguori [~anthony@pixpat.austin.ibm.com] has joined #xen
13:35	\|-\|	jerone [~jerone@pixpat.austin.ibm.com] has quit [Quit: Leaving]
13:41	\|-\|	rpg [~rpg@pixpat.austin.ibm.com] has quit [Ping timeout: 480 seconds]
13:41	\|-\|	johnlev [~johnlev@nwkea-socks-1.sun.com] has quit [Ping timeout: 480 seconds]
13:54	\|-\|	jimix [~jimix@23.94.233.220.exetel.com.au] has joined #xen
13:55	\|-\|	rpeg [~rpg@cpe-70-113-71-85.austin.res.rr.com] has joined #xen
14:27	\|-\|	aliguori [~anthony@pixpat.austin.ibm.com] has quit [Quit: Ex-Chat]
14:46	\|-\|	gerrit [~gerrit@c-67-160-130-59.hsd1.wa.comcast.net] has quit [Ping timeout: 480 seconds]
14:51	\|-\|	jimix [~jimix@23.94.233.220.exetel.com.au] has quit [Quit: jimix]
14:53	\|-\|	Sir_Ahzz [~ahzz@c-24-0-241-237.hsd1.tx.comcast.net] has quit [Remote host closed the connection]
14:54	\|-\|	bernarde [~bernarde@32.104.18.240] has quit [Ping timeout: 480 seconds]
15:00	\|-\|	Basic_java [~Basic@66.173.4.234] has quit [Quit: Leaving]
15:10	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has quit [Ping timeout: 480 seconds]
15:18	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has joined #xen
15:23	\|-\|	gerrit [~gerrit@bi01p1.co.us.ibm.com] has joined #xen
15:24	\|-\|	mcr [~mcr@CPE0006b123a026-CM0011aea1b6fc.cpe.net.cable.rogers.com] has quit [Remote host closed the connection]
15:27	\|-\|	rpeg [~rpg@cpe-70-113-71-85.austin.res.rr.com] has quit [Ping timeout: 480 seconds]
15:47	\|-\|	Basic_py [~Basic@gatekeeper.real-time.com] has joined #xen
15:53	<ns>	does anyone know if save/restore is supposed to work in 3.0?
15:54	<ns>	or if anything got fixed in there recently?
15:55	\|-\|	rpeg [~rpg@cpe-70-113-71-85.austin.res.rr.com] has joined #xen
16:10	<DV>	works in my limited testing
16:25	<sdague>	ns: there are xm tests for it, it passes those
16:26	<sdague>	if you are seeing something else fail, you should make a test case to expose it ;)
16:41	<ns>	sdague: apparently we already have tests :), this is a test case that is failing, hence the question
16:44	\|-\|	yorkedork [~thom@cpe-066-057-227-097.nc.res.rr.com] has joined #xen
16:49	\|-\|	sdague [~sdague@serenity.dague.org] has quit [Remote host closed the connection]
16:49	\|-\|	sdague [~sdague@serenity.dague.org] has joined #xen
16:53	<yorkedork>	Would anyone mind helping me figure out how I can get dom0_ops.c compiled with the kernel (it doesn't appear to be contained within the linux-2.6.merge.hg repos.)?
16:54	<yorkedork>	(or, alternatively, how to configure static or dynamic linking)
16:55	<rharper>	dom0_ops.c is part of the hypervisor, not the Xen-linux kernel
16:56	<yorkedork>	oh, i understand.
16:56	<yorkedork>	i'm trying to make a call to 'do_dom0_op(...)' from within the kernel - is that perhaps poor form?
16:57	<yorkedork>	(and consquently, i'll need to link against it while building the kernel if i continue with this method)
16:57	\|-\|	johnlev [~johnlev@nwkea-socks-2.sun.com] has joined #xen
16:57	<rharper>	you will probably have a much easier time writing a userspace app which links to libxc which provides an interface into dom0 ops
16:57	<yorkedork>	(dom0_op.h is included under xen/interface ... so i imagine it was done with linking in mind?)
16:58	<rharper>	tools/python/xen/lowlevel/xc/xc.c is one such user of libxc
16:58	<yorkedork>	well, i'm porting the powernow-k8 driver to work under xen - and i need to replace rdmsr(...) and wrmsr(...) calls with a dom0_op to modify the msrs on CPU
16:59	<rharper>	why not just make a new hypercall rather than dom0 op ?
16:59	<yorkedork>	register a hypercall to perform the function?
17:00	<rharper>	yorkedork: better yet, ask xen-devel where it should go
17:05	<tonyb>	yorkedork: If you need to do a dom0_op from a kernel you need to use the hypercal linterface.
17:05	<yorkedork>	yeah, i've just discovered my misunderstandings :)
17:06	<yorkedork>	are dom0_ops basically convenient (oft-used) wrappers for hypercalls?
17:06	<yorkedork>	built for use with user-space apps (constructing a domain, destroying, etc., etc.?)
17:06	<tonyb>	static int blah(void)
17:06	<tonyb>	{
17:06	<tonyb>	int err;
17:06	<tonyb>	dom0_op_t op = { .cmd = DOM0_BLAH,
17:06	<tonyb>	.interface_version = DOM0_INTERFACE_VERSION,
17:06	<tonyb>	.u.uniontype.arg = 1 };
17:06	<tonyb>	err = HYPERVISOR_dom0_op(&op);
17:07	<tonyb>	BUG_ON(IS_ERR((void*)err));
17:07	<tonyb>	return err;
17:07	<tonyb>	}
17:07	<tonyb>	yorkedork: there are operations that /need/ to be done in a priviledged domain, so you ask use a hypercall to make that happen.
17:08	<tonyb>	yorkedork: like futzing the pagetables or direct hardware access
17:08	<yorkedork>	yes, the analog of a system call --- i had just seen that modifying msrs was exposed through dom0_ops and I got excited that there would be less work that that.
17:09	<tonyb>	well HCALLS aren't that much work.
17:10	<tonyb>	you can even do them from userspace via the privcmd interface
17:10	<yorkedork>	hmm ... xen/interface/xen.h is the relevant file?
17:12	<yorkedork>	(I don't mean to be frustrating ... I can't seem to find docs with enough detail for me)
17:13	<tonyb>	Ummm,
17:13	<tonyb>	./linux-2.6-xen-sparse/drivers/xen/privcmd/privcmd.c
17:13	<tonyb>	./linux-2.6-xen-sparse/include/asm-xen/linux-public/privcmd.h
17:13	<yorkedork>	ah, no - for making a hypercall?
17:14	<yorkedork>	(and for reference, i am using the merge-tree ... if that makes a difference w.r.t. file hierarchy)
17:14	<tonyb>	essentially you perform an IOCTL on /proc/xen/privcmd, and it will get the kernel to do a HCALL on your behalf
17:14	<tonyb>	sorry meeting bbiaw
17:14	<yorkedork>	okay, thank you
17:15	<tab>	yorkedork: you should not be using the merge tree
17:15	<yorkedork>	why do say that?
17:16	<yorkedork>	(i am ignorant of possible reasons against this)
17:16	<tab>	because there's no reasons to use it :)
17:16	<yorkedork>	it seems more intuitive to me that the sparse tree method
17:16	<yorkedork>	and it was my understanding the sparse tree method was going to be deprecated.
17:16	<yorkedork>	perhaps i'm out in left-field :)
17:17	<tab>	the merge tree is only for people doing anything related to the linux merge
17:17	<tab>	you don't have to work in the sparse tree in xen-unstable anyway, since you work in the build directory
17:19	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has quit [Ping timeout: 480 seconds]
17:19	<yorkedork>	you changes are overwritten with each build, correct?
17:19	<yorkedork>	*your
17:19	<tab>	no
17:20	<yorkedork>	ah, okay - i feel obligated, and you've sold me on the idea :)
17:20	<tab>	you save yourself lots of trouble using xen-unstable
17:21	\|-\|	ionel [~ionel@24.87.64.156] has joined #xen
17:22	\|-\|	ionel [~ionel@24.87.64.156] has quit [Quit: ]
17:23	<yorkedork>	okay. well, thank you very much. i'm leaving range of wireless - but i'll be around later to further discuss issues after i try using hypercalls to do what i desire.
17:23	\|-\|	yorkedork [~thom@cpe-066-057-227-097.nc.res.rr.com] has quit [Quit: Leaving]
17:24	\|-\|	johnny43 [~ionel@24.87.64.156] has joined #xen
17:24	<tonyb>	yorkedork: Cool I'm around most days....
17:25	<johnny43>	guys, i'm trying to run iptables in a CentOS 4.1 / 2.6.12.6-xenU, but i get this error:
17:25	<johnny43>	Applying iptables firewall rules: iptables-restore v1.2.11: iptables-restore: unable to initializetable 'filter'
17:26	<johnny43>	iptables seem to be available in dom0.. is there anything i need to do to have it in domU?
17:27	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has joined #xen
17:29	<rharper>	domU kernel have iptable enabled in kernel config?
17:30	<johnny43>	how do i check that? i installed the xen-3.0.1 binaries
17:31	<rharper>	I think the binaries have a lot of stuff turned on, not quite sure what the config option is, lemme look
17:32	<johnny43>	thanks a bunch
17:33	<rharper>	the option should be CONFIG_IP_NF_IPTABLES, check in /boot for config-2.6.12-6-xenu on dom0 filesystem
17:33	<johnny43>	lsmod \| grep iptable shows it only in dom0 :(
17:34	<johnny43>	CONFIG_IP_NF_IPTABLES is only in the config-2.6.12.6-dom0, not in domU :(
17:34	<johnny43>	so i guess a rebuild is in order
17:35	<rharper>	seems like it
17:35	<johnny43>	thanks again for your help!
17:35	<rharper>	sure
17:36	<johnny43>	one more question...
17:36	<johnny43>	how do i enable CONFIG_IP_NF_IPTABLES specifically for domU at build-time?
17:36	<johnny43>	the instructions i've seen so far say to do make ARCH=xen in the linux-2.6.12-xen subdir...
17:36	<rharper>	you going to build xen-unstable?
17:37	<johnny43>	no, stable
17:38	\|-\|	johnlev [~johnlev@nwkea-socks-2.sun.com] has quit [Ping timeout: 480 seconds]
17:38	\|-\|	Shaun [~ndci@ip68-5-63-223.oc.oc.cox.net] has joined #xen
17:38	<rharper>	the default configs should be in linux-2.6-xen-sparse/arch/xen/configs
17:39	<rharper>	or you can cd to the kernel dir, and do as you mentioned, make ARCH=xen menuconfig, and set the option from there
17:41	<johnny43>	kernel dir == linux-2.6-xenU ?
17:43	<rharper>	johnny43: yep
17:43	<rharper>	should be linux-2.6.12-6-xenU or something like that
17:43	<johnny43>	okay, i'll try that
17:43	<Shaun>	what you trying to do?
17:44	<johnny43>	get iptables in domU
17:44	<johnny43>	aparently the xen-3.0 binaries only have it enabled in dom0
17:44	\|-\|	Basic_py [~Basic@gatekeeper.real-time.com] has quit [Quit: Leaving]
17:46	<Shaun>	so you want to go into config mode?
17:46	<Shaun>	make linux-2.6-xenU-config CONFIGMODE=menuconfig <-- i beleive is what your looking for.
17:48	<johnny43>	yes, thanks;
17:48	<Shaun>	then... make linux-2.6-xenU-build
17:48	<Shaun>	and.. make linux-2.6-xenU-install
17:49	<johnny43>	hmm.. can i skinp re-building the dom0 kernel?
17:49	\|-\|	AlexMBas [~Alexandre@200.103.241.137] has joined #xen
17:49	<Shaun>	sure
17:49	<Shaun>	what i gave you should only build domU
17:50	<Shaun>	if you wanted to do dom0 you'd use make linux-2.6-xen0-config CONFIGMODE=menuconfig and so on :)
17:50	<johnny43>	great, thanks
17:51	<johnny43>	now how do i tell build to use PAE? make XEN_TARGET_X86_PAE=y linux-2.6-xenU-build ?
17:52	<Shaun>	append XEN_TARGET_X86_PAE=y
17:52	<Shaun>	i usually do `export XEN_TARGET_X86_PAE=y`
17:52	<Shaun>	just so i make sure i dont forget it :)
17:52	<johnny43>	ok, thanks
17:59	\|-\|	woody [~woody@bi01p1.co.us.ibm.com] has joined #xen
18:02	\|-\|	rharper [~rharper@pixpat.austin.ibm.com] has quit [Quit: Leaving]
18:05	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has quit [Quit: Leaving]
18:05	\|-\|	dansmith [~dan@bi01p1.co.us.ibm.com] has quit [Quit: Leaving]
18:06	\|-\|	johnlev [~johnlev@nwkea-socks-2.sun.com] has joined #xen
18:08	\|-\|	ivan [~ikelly@83-71-24-225.b-ras1.lmk.limerick.eircom.net] has joined #xen
18:09	<ivan>	how important is the mfn_list passed in through start_info_t?
18:16	\|-\|	jdmason [~jonmason@pixpat.austin.ibm.com] has quit [Remote host closed the connection]
18:16	\|-\|	jdmason [~jonmason@pixpat.austin.ibm.com] has joined #xen
18:30	<johnny43>	thanks guys - i now have a happy iptables-enabled domU :)
18:32	\|-\|	ivan [~ikelly@83-71-24-225.b-ras1.lmk.limerick.eircom.net] has quit [Quit: leaving]
18:33	<Shaun>	johnny43: i cant beleive it's not enabled by default, guess i'm going to have to run through my config before i go live... didnt even think that it wouldnt be included...
18:33	\|-\|	dansmith [~dan@c-24-21-32-166.hsd1.or.comcast.net] has joined #xen
18:35	<johnny43>	well.. i launched menuconfigs for domU and dom0, and they sure weren't enabled for domU
18:35	<johnny43>	i enabled in domU everything that was enagled in dom0
18:44	\|-\|	Basic_py [~Basic@warden.real-time.com] has joined #xen
19:08	\|-\|	pvanhoof [~pvanhoof@d54C0FBBD.access.telenet.be] has quit [Remote host closed the connection]
19:16	\|-\|	aliguori [~anthony@cpe-70-116-13-229.austin.res.rr.com] has joined #xen
19:30	\|-\|	probonic [~probonic@host-84-9-223-181.bulldogdsl.com] has joined #xen
19:31	\|-\|	ns [~niv@bi01p1.co.us.ibm.com] has quit [Quit: Leaving]
19:31	<probonic>	hey everyone
19:31	<probonic>	I have a quick question -
19:32	<probonic>	"xm mem-set" doesn't seem to be having any effect on Domain 0. When I do "xm mem-set Domain-0 640", top and free both still show 192MB
19:32	<probonic>	I am running Xen 3.0.0
19:32	<probonic>	any ideas?
19:45	<johnlev>	anyone know the minimum python version needed by xm?
19:50	\|-\|	gerrit [~gerrit@bi01p1.co.us.ibm.com] has quit [Ping timeout: 480 seconds]
19:51	<aliguori>	johnlev, i'm pretty sure it's 2.3
19:51	<johnlev>	great.
19:51	<aliguori>	there's been some effort to make it work on 2.2 but i doubt it still does
19:51	<aliguori>	you may be able to get it to work on 2.2 with sufficient hacking
19:52	\|-\|	sonic [~sonic@83.141.118.40] has joined #xen
19:52	<sonic>	Hello?
19:53	<johnlev>	aliguori: I want 2.3 anyway
19:58	<Shaun>	aliguori: with UML it was a problem security issue to enable guest kernels with module support... is this the same with domU?
20:00	<aliguori>	Shaun, nope
20:06	\|-\|	rusty [~rusty@bh02i525f01.au.ibm.com] has joined #xen
20:14	<sonic>	Hello, can anyone help me please?
20:16	\|-\|	woody [~woody@bi01p1.co.us.ibm.com] has quit [Remote host closed the connection]
20:20	\|-\|	woody [~woody@bi01p1.co.us.ibm.com] has joined #xen
20:22	<sonic>	Is there extra to do when installing xen on a RAIDed system?
20:24	\|-\|	dansmith [~dan@c-24-21-32-166.hsd1.or.comcast.net] has quit [Quit: Leaving]
20:42	<sonic>	Can anyone help me, I'm having serious problems.
20:48	<aliguori>	sonic, are you using an initrd?
20:48	<aliguori>	that you built specifically for xen
20:48	<sonic>	no, I don't know how to do that
20:48	<aliguori>	sonic, that's your problem
20:48	<aliguori>	read the FAQ
20:48	<sonic>	I'm on a RAIDed system, are there problems with that?
20:48	<aliguori>	no, you need an initrd
20:49	<sonic>	Ok, I'll have a read, Cheers!
20:49	<sonic>	Also, I'm on a k6II processor. A friend recons there could be issues with that. What do you think?
20:49	<aliguori>	sonic, come back if you have problems, i really want to make sure the FAQ is a good source for help so it's useful to know if it's helpful
20:49	<aliguori>	no, that should be fine
20:50	<sonic>	aliguori: Thanks, I'll have a read and come back!
20:50	<aliguori>	np
20:50	<aliguori>	good luck
20:50	<sonic>	Cheers!
20:50	<sonic>	Will you be here for a while?
20:50	<aliguori>	sonic, i'm always here :-)
20:52	<sonic>	Cool, be back shortly!
20:57	<Shaun>	is their a way to see free ram that has not been used by dom0/U?
20:57	<Shaun>	basically so i can figure out how much is left :)
20:57	<aliguori>	Shaun, xm info
20:58	<Shaun>	thanks
20:59	<sonic>	aliguori: The FAQ seems to assume knowledge of initrd files. Do I have to create one in the boot directory first, or does the GRUB command do this for me?
21:00	<aliguori>	sonic, where you looking at the wiki faq?
21:00	<sonic>	http://wiki.xensource.com/xenwiki/XenFaq
21:01	<aliguori>	ok, i see, it sucks wrt to initrds
21:01	<aliguori>	what distro are you on?
21:01	<sonic>	Debian
21:01	<aliguori>	ah, okay, you want to use a command like this:
21:01	<sonic>	I can see an initrd file in my boot directory
21:01	<rusty>	aliguori: got bench results for shareblk.
21:02	<aliguori>	mkinitrd -o /boot/initrd.img-2.6.12-xen0 2.6.12-xen0
21:02	<aliguori>	replacing 2.6.12 with the appropriate xen version
21:02	<aliguori>	then, in your grub.conf, make sure it uses that initrd (but use the module directive instead of the initrd directive)
21:02	<aliguori>	rusty, ?????
21:02	<aliguori>	drum roll
21:02	<sonic>	aha, like 2.6.12.11?
21:02	<aliguori>	sonic, yeah, if that's the version of your xen kernel
21:02	<aliguori>	look in /boot and see what kernel has a -xen0
21:03	<aliguori>	and use that version
21:03	<aliguori>	(including the -xen0)
21:03	<sonic>	Cheers, I'll give it a try and report back, thanks!
21:03	<aliguori>	np
21:04	\|-\|	cods [~cods@tuxee.net] has joined #xen
21:05	\|-\|	hollisb [~hollisb@user-0vvde2g.cable.mindspring.com] has joined #xen
21:05	<rusty>	Their results: dd bs=1M count=1000 71.5 seconds
21:06	<aliguori>	k
21:06	<sonic>	aliguori: I tried that first command and got the following errors:
21:06	<sonic>	mkinitrd -o /boot/initrd.img-2.6.11.12-xen0 2.6.11.12-xen0
21:07	<sonic>	/usr/sbin/mkinitrd: add_modules_dep_2_5: modprobe failed
21:07	<rusty>	Their results: dbench 10: 48.685 MB/sec
21:07	<sonic>	FATAL: Module raid0 not found.
21:07	<sonic>	FATAL: Module raid1 not found.
21:07	<sonic>	FATAL: Module quota_v1 not found.
21:07	<sonic>	FATAL: Module quota_v2 not found.
21:07	<sonic>	WARNING: This failure MAY indicate that your kernel will not boot!
21:07	<sonic>	but it can also be triggered by needed modules being compiled into
21:07	<sonic>	the kernel.
21:07	<rusty>	My results: dd: 31.1 seconds. dbench 10: 106.997 MB/sec
21:07	<aliguori>	sonic, huh, and 2.6.11.12 is the right kernel? ...
21:08	<aliguori>	rusty, holy crap!
21:08	<sonic>	Yep
21:08	<aliguori>	rusty, and this is a userspace driver???
21:08	<aliguori>	wow
21:08	<aliguori>	sonic, xen hasn't used that kernel in a very long time..
21:08	<aliguori>	oh
21:08	<aliguori>	you're using "a perfect xen setup"?
21:08	<sonic>	Yep, from howto forge
21:08	<sonic>	Compiled from source
21:08	<aliguori>	i highly recommend just doing a make install and not following those instructions
21:08	<aliguori>	it's a bit insane for users to be patching distro sources
21:09	<sonic>	Is there a tutorial you recommend?
21:09	<aliguori>	rusty, what do you attribute the speed up to?
21:09	<rusty>	aliguori: Yes, userspace. Means I'm copying everything twice.
21:09	<aliguori>	sonic, just make install... you don't really need a tutorial
21:09	<rusty>	aliguori: I am not fast. They are unforgivably slow.
21:09	<aliguori>	why does xenblk suck so much?
21:11	<rusty>	aliguori: I don't actually know. It's hard to attribute that much damage to the overhead of page table manips.
21:11	<sonic>	Since I've followed that tutorial then should I delete everything and download another version of the xen source? Wouldn't a make install just use my modified menuconf?
21:11	<aliguori>	rusty, the block drivers don't do page flipping either
21:12	<rusty>	aliguori: they don't grant pages to the other domain??
21:12	<aliguori>	sonic, yeah, grab a new version (3.0.1) and install that from scratch
21:12	<aliguori>	rusty, they don't page flip, i don't know what the mechanism is.. one sec, easy enough to determine
21:14	<aliguori>	rusty, okay, they do grant access so there is page table modification.. just not page flipping per say
21:14	<sonic>	aliguori: Does 3.0.1 have quotas, iptables, and dummy network drivers enabled by default?
21:14	<hollisb>	they DMA directly into the target domain's mfn
21:14	<rusty>	aliguori, hollisb: that's what I thought (welcome back, Hollis!)
21:14	<hollisb>	hi :)
21:15	<hollisb>	right now I'm attempting to stay awake ;)
21:15	<aliguori>	heh
21:15	<aliguori>	hollisb, are you back in the states?
21:15	<hollisb>	yep. I hope to be in the office tomorrow
21:15	<aliguori>	cool
21:15	<hollisb>	(but notice I haven't mentioned a time ;)
21:15	<aliguori>	hehe
21:16	<aliguori>	rusty, are you on a UP or SMP system?
21:16	<rusty>	aliguori: UP. I asked niv to do some SMP testing, but havent' seen her to harass her again.
21:17	<aliguori>	rusty, I think they batch up the requests, if you're doing a event channel notification for every bio, that might explain it...
21:18	<rusty>	aliguori: ON the dd test????
21:18	<aliguori>	rusty, yeah, i can't explain that...
21:18	<rusty>	aliguori: me neither.
21:19	<sonic>	aliguori: I might leave the 3.0.1 install till the morning and get a nights sleep first. Thanks very much for your time! I'll check in here again tomorrow night and let you know how I got on!
21:19	<aliguori>	rusty, you should see if andrew theurer can do some performance testing of it.. he's got a pretty good idea of what the current bottle necks are
21:19	<aliguori>	sonic, ok :-)
21:19	<aliguori>	rusty, he's got big systems too...
21:20	<sonic>	Thanks dudes, signing off........
21:20	\|-\|	sonic [~sonic@83.141.118.40] has left #xen [Konversation terminated!]
21:23	<rusty>	aliguori: hmm, if you want to ping him that would be good. It's still in the manual-configuration crashes-on-close state at the moment tho 8)
21:24	<aliguori>	rusty, ah :-)
21:24	<aliguori>	rusty, is everything in hg?
21:32	<rusty>	aliguori: Yes, just writing release notes now.
21:43	\|-\|	pdx6 [~ballew@screen.sublinear.net] has quit [Remote host closed the connection]
21:56	\|-\|	aliguori [~anthony@cpe-70-116-13-229.austin.res.rr.com] has quit [Remote host closed the connection]
21:57	\|-\|	aliguori [~anthony@cpe-70-116-13-229.austin.res.rr.com] has joined #xen
22:07	\|-\|	probonic [~probonic@host-84-9-223-181.bulldogdsl.com] has quit [Ping timeout: 480 seconds]
22:08	\|-\|	pdx6 [~ballew@screen.sublinear.net] has joined #xen
22:11	<hollisb>	aliguori: so vm-tools doesn't have any sort of daemon in it? they're all run-once tools?
22:11	<hollisb>	hmm, there is a xenctld..
22:14	<aliguori>	hollisb, yeah, but i always hated that
22:14	<aliguori>	hollisb, the earlier versions had no daemons
22:14	<aliguori>	hollisb, i had to add xenctld though to support domain reboots
22:14	<hollisb>	hmm, I haven't needed one for what I've been doing so far
22:15	<aliguori>	hollisb, yeah, so, i was in the same exact position.. but once they got useful enough, reboot handling became necessary
22:15	<hollisb>	ok. so far you haven't talked me out of seeing if I can just use vm-tools :)
22:15	<aliguori>	hollisb, okay.. and for development, i think it's just fine
22:15	<hollisb>	ok
22:16	<aliguori>	in fact, it'll probably be worlds easier
22:16	<aliguori>	jacob posted a link with a more current domain builder too that used the store properly too btw
22:16	<aliguori>	when i say domain builder, i mean something to create domains...
22:16	<aliguori>	not a libxc builder
22:16	<aliguori>	:-)
22:17	<hollisb>	ok. the thing that scares me about Jacob's stuff though is that he wants in-domain code to do the bootstrapping
22:17	\|-\|	pdx6 [~ballew@screen.sublinear.net] has quit [Quit: forgot to load screen first]
22:17	<aliguori>	hehhe, and migration :-)
22:17	<hollisb>	uh huh :)
22:19	\|-\|	pdx6 [~ballew@screen.sublinear.net] has joined #xen
22:19	\|-\|	pdx6 [~ballew@screen.sublinear.net] has quit [Quit: ]
22:20	\|-\|	pdx6 [~ballew@screen.sublinear.net] has joined #xen
22:29	<anball>	so, does anyone know where /local/domain/<domid> gets filled out on a domU's creation? i'm lookin in tools/python/xend/XendDomainInfo.py and nearby ...
22:32	<anball>	ah, i figured out what i needed to know without learning this, but would still like to know
22:32	\|-\|	yorkedork [~thom@cpe-066-057-227-097.nc.res.rr.com] has joined #xen
22:40	\|-\|	gerrit [~gerrit@c-67-160-130-59.hsd1.wa.comcast.net] has joined #xen
22:49	<aliguori>	anball, where the directory gets created?
22:51	<aliguori>	anball, xenstored actually creates /local/domain/<domid>....
22:52	<aliguori>	xend fills out some of the information in it too though
22:53	<aliguori>	or maybe not :-)
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has left #xen [Rotating Logs]
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
23:01	<rusty>	Is the xen-devel list backed up? I haven't seen my [BUNDLE] post come thru...
23:13	<anball>	aliguori: thanks. didn't know you were still up. I found what i needed in /local/domain/<domid>/vm . there is no /local/domain/<domid>/uuid , but that's okay.
23:40	<hollisb>	rusty: I see your mail I think
23:42	<rusty>	hollisb: thx
23:53	\|-\|	rpeg [~rpg@cpe-70-113-71-85.austin.res.rr.com] has quit [Remote host closed the connection]
---	Log	closed Fri Feb 17 00:00:14 2006