TheGrebs.com:IRC Logs:#xen:2006:02:27

---	Log	opened Mon Feb 27 00:00:52 2006
00:11	\|-\|	hormsbert [~horms@vagw.valinux.co.jp] has joined #xen
00:14	\|-\|	hormsbert [~horms@vagw.valinux.co.jp] has quit [Quit: ]
00:15	\|-\|	hormsbert [~horms@vagw.valinux.co.jp] has joined #xen
00:17	\|-\|	hormsbert [~horms@vagw.valinux.co.jp] has quit [Quit: ]
00:18	\|-\|	horms [~horms@vagw.valinux.co.jp] has quit [Ping timeout: 480 seconds]
00:43	\|-\|	Netsplit arion.oftc.net <-> oxygen.oftc.net quits: spb, michal`, Shaun, JViz, andy
00:43	\|-\|	pdx6 [~ballew@209.237.242.37] has quit [Read error: Connection reset by peer]
00:43	\|-\|	pdx6 [~ballew@209.237.242.37] has joined #xen
00:43	\|-\|	Netsplit arion.oftc.net <-> quasar.oftc.net quits: VS_ChanLog
00:43	\|-\|	Netsplit over, joins: Shaun, JViz, spb, michal`, andy
00:43	\|-\|	JViz [Anomaly@adsl-065-013-131-023.sip.int.bellsouth.net] has quit [Read error: Connection reset by peer]
00:44	\|-\|	JViz [Anomaly@adsl-065-013-131-023.sip.int.bellsouth.net] has joined #xen
00:44	\|-\|	Netsplit over, joins: VS_ChanLog
00:50	\|-\|	horms [~horms@vagw.valinux.co.jp] has joined #xen
02:37	\|-\|	sdog [~sdog@62.58.98.210] has joined #xen
03:19	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
03:31	\|-\|	athomas [~athomas@host86-136-81-69.range86-136.btcentralplus.com] has joined #xen
03:46	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has quit [Quit: Leaving]
03:48	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
03:48	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has quit [Remote host closed the connection]
03:51	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
04:03	\|-\|	Xgemini\|Soul_lost [~eXtreme@par69-1-82-67-25-110.fbx.proxad.net] has joined #xen
04:03	<Xgemini\|Soul_lost>	hello all
04:06	<Xgemini\|Soul_lost>	j'ai comme un p'tit soucis suite a une recompilation de mon noyau ...un p'tit message a répétition du type "Currently emulating unsupported memory accesses ..."
04:08	<Xgemini\|Soul_lost>	j'ai déjà fait un "mv /lib/tls /lib/tls.disabled " ...pourtant le problème continue ... quelqu'un aurait une p'tit idée ... ?? Merci d'avance
04:10	<hensema>	Xgemini\|Soul_lost: have you rebooted the virtual machine? The TLS libraries remain open (and therefore active) after renaming them
04:16	\|-\|	horms [~horms@vagw.valinux.co.jp] has quit [Quit: Leaving]
04:17	\|-\|	Xg [~axrezo@par69-1-82-67-25-110.fbx.proxad.net] has joined #xen
04:17	\|-\|	Xg [~axrezo@par69-1-82-67-25-110.fbx.proxad.net] has quit [Quit: ]
04:17	\|-\|	Xg [~axrezo@par69-1-82-67-25-110.fbx.proxad.net] has joined #xen
04:18	\|-\|	Xg [~axrezo@par69-1-82-67-25-110.fbx.proxad.net] has quit [Quit: ]
04:21	<mancdaz>	how would I get a SuSE 10 guest on a fedora 5 host system?
04:30	<hensema>	mancdaz: http://dexter.hensema.net/~erik/xen/
04:30	<hensema>	suse xen kernel and suse install image on initrd
04:30	\|-\|	Tv [~tv@i1.inoi.fi] has joined #xen
04:34	<mancdaz>	excellent thanks but excuse my newbiewness - how would I use them to get suse installed?
04:37	<hensema>	just boot it and the installer will start, select network install and off you go ;-)
04:37	<hensema>	kernel = "/var/lib/xen/images/domain1/vmlinuz-2.6.13-15-xen"
04:37	<hensema>	ramdisk = "/var/lib/xen/images/domain1/inst-initrd"
04:42	<mancdaz>	hmm I am getting a kernel panic when booting that kernel
04:42	<hensema>	what's the panic?
04:43	<mancdaz>	VFS: Cannot open root device "<NULL>" or unknown-block(0,0)
04:43	<mancdaz>	Please append a correct "root=" boot option
04:43	<mancdaz>	Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(0,0)
04:43	<mancdaz>	I am using an lvm partition
04:43	<mancdaz>	as the target for the guest
04:46	<hensema>	hmmm
04:46	<hensema>	try passing it a root=/dev/hda1 parameter then
04:48	<mancdaz>	does that get passed in the xen configuration file for that guest, in the kernel = "path/to/kernel" bit?
05:11	\|-\|	horms [~horms@p2191-ipbf409marunouchi.tokyo.ocn.ne.jp] has joined #xen
05:26	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
05:48	\|-\|	Tv [~tv@i1.inoi.fi] has quit [Quit: foo]
05:55	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has quit [Quit: mdday]
06:09	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
06:09	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has quit [Quit: ]
06:10	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
06:10	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has quit [Quit: ]
06:11	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
06:17	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has quit [Quit: mdday]
06:32	\|-\|	Tv [~tv@GMMDXXVII.dsl.saunalahti.fi] has joined #xen
06:52	\|-\|	AlexMBas [~Alexandre@200.138.34.244] has joined #xen
06:55	\|-\|	AlexMBas [~Alexandre@200.138.34.244] has quit [Remote host closed the connection]
07:10	\|-\|	AlexMBas [~Alexandre@200.138.34.244] has joined #xen
07:15	\|-\|	Xgemini\|Soul_lost [~eXtreme@par69-1-82-67-25-110.fbx.proxad.net] has quit [Quit: Secrêt caché du DOS : ajoutez BUGS=OFF sur votre fichier CONFIG.SYS]
07:19	\|-\|	ivan [~ikelly@83-71-74-95.b-ras1.lmk.limerick.eircom.net] has joined #xen
07:24	\|-\|	Ghosty [~Ghosty@s5590c066.adsl.wanadoo.nl] has joined #xen
07:25	<Ghosty>	hi guys can someone help me i am having troubles with Fedora 4 and Xen with GRUB
07:25	<Ghosty>	i was using the automatic install but grub seems to have a problem with displaying xen in the menu
07:27	<riel>	oh ? how so ?
07:27	<Ghosty>	it is not showing up in the list but its there in the grub.conf
07:28	<Ghosty>	and also menu.lst which is a symbolic link i believe
07:28	\|-\|	horms [~horms@p2191-ipbf409marunouchi.tokyo.ocn.ne.jp] has quit [Quit: Leaving]
07:31	<Ghosty>	i am using RAID 1 but this can maybey the problem because i have just changed the name to **** me
07:32	<Ghosty>	and it still shows the same
07:33	<riel>	are you sure raid 1 is working fine for /boot too ?
07:33	<riel>	I wonder if you're writing to the /boot on the disk that the BIOS isn't reading ;)
07:37	<Ghosty>	i think so 2 :\|
07:37	<Ghosty>	any suggestions :)
07:37	<Ghosty>	the part of RAID 1 is new to me but i got the good how to /// to get the job done
07:39	<Ghosty>	and the server reboots again :)
07:40	<riel>	cat /proc/mounts after booting
07:40	<riel>	then check the raid setup, to make sure /boot really is raid and not just /
07:41	<Ghosty>	ok booting now gonna check in a few secs
07:42	<Ghosty>	-------------/dev/md2 /boot ext3 rw 0 0
07:42	<Ghosty>	i guess i think so
07:47	\|-\|	johnlev [~johnlev@nwkea-socks-1.sun.com] has joined #xen
07:47	<Ghosty>	are there other ways to check if your RAID 1 is properly conf'ed
07:55	\|-\|	muli_ [~muli@nesher3.haifa.il.ibm.com] has joined #xen
08:24	\|-\|	mancdaz [~mancdaz@blueice2n1.uk.ibm.com] has quit [Quit: Leaving]
08:24	<Ghosty>	i am now fixing this raid 1 problem i am having and now i have changed and made the swap and boot partition good
08:25	<Ghosty>	but now GRUB hangs on reboot
08:25	\|-\|	riel changed nick to surriel
08:40	\|-\|	anball [~anball@bi01p1.nc.us.ibm.com] has joined #xen
08:49	\|-\|	Basic_py [~Basic@warden.real-time.com] has quit [Quit: Leaving]
08:49	\|-\|	Netsplit unununium.oftc.net <-> europa.oftc.net quits: surriel, Hunger
08:54	\|-\|	Netsplit over, joins: Hunger, surriel
09:00	\|-\|	mdday [~mdday@bi01p1.nc.us.ibm.com] has joined #xen
09:03	\|-\|	dansmith [~dan@129.33.1.37] has joined #xen
09:05	\|-\|	unriel changed nick to riel
09:08	\|-\|	jimix [~jimix@129.33.1.37] has joined #xen
09:09	\|-\|	harry_ [~harry@blueice4n2.uk.ibm.com] has joined #xen
09:09	<harry_>	hi all
09:09	\|-\|	jimix [~jimix@129.33.1.37] has quit [Quit: ]
09:09	<harry_>	save - restore working in current unstable?
09:12	\|-\|	rharper [~rharper@pixpat.austin.ibm.com] has joined #xen
09:12	\|-\|	muli_ [~muli@nesher3.haifa.il.ibm.com] has quit [Quit: My damn controlling terminal disappeared!]
09:18	<mikegrb>\|-	muli_ [~muli@nesher3.haifa.il.ibm.com] has quit [Quit: My damn controlling terminal disappeared!]
09:18	<mikegrb>	[09:18][09:18][mikegrb(+Ri)][7:#xen(+ns)][Act: 3,4,5,6,9] [#xen] \|-\| muli_ [~muli@nesher3.haifa.il.ibm.com] has quit [Quit: My damn controlling terminal disappeared!]
09:18	<mikegrb>	[09:18][09:18][mikegrb(+Ri)][7:#xen(+ns)][Act: 3,4,5,6,9] [#xen] \|-\| muli_ [~muli@nesher3.haifa.il.ibm.com] has quit [Quit: My damn controlling terminal disappeared!]
09:18	<mikegrb>	[09:18][09:18][mikegrb(+Ri)][7:#xen(+ns)][Act: 3,4,5,6,9] [#xen] \|-\| muli_ [~muli@nesher3.haifa.il.ibm.com] has quit [Quit: My damn controlling terminal disappeared!]
09:18	<mikegrb>	[09:18][09:18][mikegrb(+Ri)][7:#xen(+ns)][Act: 3,4,5,6,9] [#xen] \|-\| muli_ [~muli@nesher3.haifa.il.ibm.com] has quit [Quit: My damn controlling terminal disappeared!]
09:18	<mikegrb>	[09:18][09:18][mikegrb(+Ri)][7:#xen(+ns)][Act: 3,4,5,6,9] [#xen] \|-\| muli_ [~muli@nesher3.haifa.il.ibm.com] has quit [Quit: My damn controlling terminal disappeared!]
09:18	<mikegrb>	[09:18][09:18][mikegrb(+Ri)][7:#xen(+ns)][Act: 3,4,5,6,9] [#xen] \|-\| muli_ [~muli@nesher3.haifa.il.ibm.com] has quit [Quit: My damn controlling terminal disappeared!]
09:19	<mikegrb>	[09:18][09:18][mikegrb(+Ri)][7:#xen(+ns)][Act: 3,4,5,6,9] [#xen] 0..0.0
09:21	<Ghosty>	i got the RAID one part working and i can boort XEN
09:21	<Ghosty>	boot xen i mean
09:21	<Ghosty>	but when i boot xen i get a KERNEL PANIC and it reboots again (man i thought this was a windows joke :P)
09:25	\|-\|	bestorga [~bestorga@63-230-161-130.ptld.qwest.net] has joined #xen
09:30	\|-\|	Ghosty [~Ghosty@s5590c066.adsl.wanadoo.nl] has quit [Quit: ]
09:32	\|-\|	schultmc_ [~schultmc@zealot.progeny.com] has joined #xen
09:32	\|-\|	schultmc_ [~schultmc@zealot.progeny.com] has quit [Read error: Connection reset by peer]
09:34	\|-\|	jimix [~jimix@129.33.1.37] has joined #xen
09:38	\|-\|	jimix [~jimix@129.33.1.37] has quit [Quit: ]
10:18	\|-\|	Ghosty [~Ghosty@s5590c066.adsl.wanadoo.nl] has joined #xen
10:18	<Ghosty>	howdy peeps
10:18	<Ghosty>	got a Question again about xen
10:19	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has joined #xen
10:19	<Ghosty>	this time i got a kernel panic i am using Fedora 4 with xen 3.0 on a dell rckserver
10:20	<Ghosty>	unable to read superblock --->> kill init then a Kernel Panic
10:20	<Ghosty>	which immediatly causes a rebooit
10:20	<Ghosty>	i am now YUM updating everything
10:20	<Ghosty>	and it is not my RAID config because i just fixed this
10:21	<Ghosty>	someone with the same problems ??
10:21	\|-\|	mancdaz [~mancdaz@blueice3n1.uk.ibm.com] has joined #xen
10:21	\|-\|	mancdaz [~mancdaz@blueice3n1.uk.ibm.com] has quit [Quit: ]
10:21	\|-\|	mancdaz [~mancdaz@blueice2n1.uk.ibm.com] has joined #xen
10:22	<Ghosty>	SOMEONE!!!!!!!!!!!!!!!!!!! :)
10:23	<Ghosty>	also i am using 2 80 GB HD'vez in RAID 1
10:26	\|-\|	hbaum [~hbaum@129.33.1.37] has joined #xen
10:28	\|-\|	Solist [~Solist@mvil2.bb.netvision.net.il] has joined #xen
10:28	<harry_>	Ghosty: the kernel panic and the unable to read superblock mean that the kernel can't read the root filesystem
10:29	<harry_>	Ghosty: which might mean that your grub config for the kernel that Xen is booting is wrong
10:30	\|-\|	cableman [~cableman@3E6B4910.rev.stofanet.dk] has joined #xen
10:33	\|-\|	sdog [~sdog@62.58.98.210] has left #xen []
10:33	<Ghosty>	mmmmm thanks harry
10:33	<Ghosty>	so what am i to do update change remove ???
10:34	<Ghosty>	. . this sucks .
10:34	<harry_>	Ghosty: is the machine booting the dom0 domain in native linux ok?
10:35	<Ghosty>	Nope
10:35	<Ghosty>	it doesnt reach that point because i get this panic and immediate reboot
10:35	<Ghosty>	or do i need to recompile my kernel on my server
10:35	\|-\|	AlexMBas [~Alexandre@200.138.34.244] has quit [Quit: Fui embora]
10:35	<harry_>	Ghosty: do you have a boot option in your grub config to boot the machine without xen?
10:36	<Ghosty>	yes
10:36	<Ghosty>	and this works
10:36	<Ghosty>	without a problem
10:36	<harry_>	Ghosty: OK so that boots your dom0 domain but under native linux not xen.
10:37	<harry_>	Ghosty: so have a look at your grub config
10:37	<Ghosty>	oooooo i get it now :) still a noob at alot of things ;)
10:38	<mancdaz>	anyone tried installing a suse 10 guest in a fc5 host?
10:39	<mancdaz>	mine keeps bombing out at when trying to run the installer via nfs
10:39	<Ghosty>	sorry i am a puzzz when it comes to BETA
10:40	\|-\|	Ghosty [~Ghosty@s5590c066.adsl.wanadoo.nl] has quit [Quit: ]
10:41	<mancdaz>	nfs mounts up ok, starts to load the install environment, then I get "lockd: failed to open /var/lib/nfs/state: err=-2"
10:41	<mancdaz>	in the backgriund
10:47	\|-\|	mejlholm [~mejlholm@port79.ds1-abc.adsl.cybercity.dk] has quit [Quit: Client exiting]
10:48	\|-\|	surriel [riel@fangorn.surriel.com] has quit [Ping timeout: 480 seconds]
10:50	\|-\|	kwarz [~root@surf.mindwerk.de] has joined #xen
10:50	<kwarz>	hi
10:50	\|-\|	surriel [riel@fangorn.surriel.com] has joined #xen
10:58	\|-\|	jerone [~jerone@adsl-71-145-138-37.dsl.austtx.sbcglobal.net] has joined #xen
11:01	<Solist>	does anybody here knows about qemu disk emulation in vt-x?
11:01	<Solist>	My question is : does it use dma ?
11:02	<Solist>	I know that when using pcnet (AMD nic emulation) you use DMA
11:02	<Solist>	pcnet is the default ; but when using ne2000
11:02	<Solist>	emulation you DON'T use DMA
11:20	\|-\|	cableman [~cableman@3E6B4910.rev.stofanet.dk] has quit [Read error: Connection reset by peer]
11:22	\|-\|	cableman [~cableman@3E6B4910.rev.stofanet.dk] has joined #xen
11:23	\|-\|	stekloff [~stekloff@129.33.1.37] has joined #xen
11:33	\|-\|	kwarz [~root@surf.mindwerk.de] has quit [Quit: BitchX: use it, it makes you bulletproof]
11:36	\|-\|	athomas [~athomas@host86-136-81-69.range86-136.btcentralplus.com] has quit [Quit: Leaving]
11:43	\|-\|	dansmith [~dan@129.33.1.37] has quit [Quit: Leaving]
11:43	\|-\|	dansmith [~dan@129.33.1.37] has joined #xen
11:46	\|-\|	harry_ [~harry@blueice4n2.uk.ibm.com] has quit [Quit: Leaving]
11:59	\|-\|	zwane [~zwane@S0106001217d33fa6.vc.shawcable.net] has quit [Quit: Leaving]
12:07	\|-\|	Solist [~Solist@mvil2.bb.netvision.net.il] has quit [Quit: Leaving]
12:15	\|-\|	muli_ [~muli@87.69.40.180.cable.012.net.il] has joined #xen
12:27	\|-\|	hollisb [~hollisb@pixpat.austin.ibm.com] has joined #xen
12:43	\|-\|	Basic_py [~Basic@mailserver.dayport.com] has joined #xen
12:59	\|-\|	gerrit [~gerrit@c-67-160-146-170.hsd1.or.comcast.net] has quit [Ping timeout: 480 seconds]
13:06	\|-\|	dansmith_ [~dan@129.33.1.37] has joined #xen
13:06	\|-\|	dansmith [~dan@129.33.1.37] has quit [Quit: Leaving]
13:10	\|-\|	secureboot [~secureboo@coriakin.cse.psu.edu] has joined #xen
13:11	<secureboot>	how do multiple domUs all get events from a device? If each interrupt signals each virq on the physical IRQ, won't all domUs get all events, and thus be able to get all data from all devices?
13:11	<secureboot>	obviously, i'm misunderstanding something basic here
13:14	<secureboot>	i guess my basic question is: when a lot of domUs are all registered to receive interrupts from a given device, how is access to the device mediated such that each domU can only see the information it should be able to see about that device?
13:15	\|-\|	f [~f@p54899015.dip0.t-ipconnect.de] has joined #xen
13:18	<tonfa>	can more than one domU receive a physical interrupt ?
13:33	\|-\|	ns [~nivedita@c-67-171-181-157.hsd1.or.comcast.net] has quit [Quit: Leaving]
13:33	\|-\|	stekloff [~stekloff@129.33.1.37] has quit [Ping timeout: 480 seconds]
13:34	\|-\|	dansmith_ [~dan@129.33.1.37] has quit [Ping timeout: 480 seconds]
13:36	\|-\|	Edgan [~edgan@64-42-67-14.atgi.net] has joined #xen
13:37	<Edgan>	Does anyone know a way to mount partitions in a logical volume from the host? As in /dev/vg/lv has xvda1 and xvda2, so it isn't just a filesystem, but subpartitions.
13:41	<f>	Edgan: either using losetup or kpartx
13:41	<f>	or manually specifying the offset to -o loop,offset=...
13:44	\|-\|	nod [jeremy@207.70.175.23] has left #xen [too many chans to watch. :)]
13:44	\|-\|	dansmith_ [~dan@129.33.1.37] has joined #xen
13:44	<Edgan>	f: yeah, loop had come to mind. Didn't think of some offset, but that sounds messy. Will looking into kpartx
13:44	\|-\|	stekloff [~stekloff@129.33.1.37] has joined #xen
13:47	<secureboot>	tonfa: from http://wiki.xensource.com/xenwiki/XenIntro, it seems that is the case
13:48	\|-\|	gerrit [~gerrit@129.33.1.37] has joined #xen
13:48	<secureboot>	_do_IRQ_guest() send the interrupt by calling send_guest_pirq() to all guests who are registered on this IRQ
13:48	<secureboot>	quoted from that page
13:52	\|-\|	bestorga [~bestorga@63-230-161-130.ptld.qwest.net] has quit [Quit: Leaving]
13:53	<f>	edgan: losetup might be actually the easiest. there must be some reason the Xen guys are shipping it :)
13:53	<Edgan>	f: ok, I will look into both when I am around my Xen machine.
13:53	<f>	erm
13:53	<f>	actually not losetup, loinit it's called
13:54	<f>	i either used kpartx or the offsets myself
13:55	<hollisb>	secureboot: do you see the irq_access_permitted check in evtchn_bind_pirq?
13:55	<Edgan>	f: :)
13:55	<hollisb>	secureboot: only dom0 is given access (via irq_permit_access)
13:56	<tonfa>	hollisb: why was it designed to permit sharing irq ?
13:56	<secureboot>	hollisb: so the hypervisor is the only one who accesses the device? then how does it figure out what data exactly can go to which domU?
13:57	<hollisb>	tonfa: because having non-dom0 "driver domains", i.e. domains with direct access to some devices, is a pretty important thing
13:57	\|-\|	aliguori [~anthony@cpe-70-116-13-229.austin.res.rr.com] has joined #xen
13:57	<hollisb>	tonfa: however, driver domain support was removed a while back and hasn't been re-added since then (AFAIK)
13:58	<hollisb>	the infrastructure patiently awaits its return I guess :)
13:58	<tonfa>	ok
13:58	<hollisb>	secureboot: your question isn't really making sense to me
13:58	<aliguori>	actually, i thought keir committed those patches recently
13:59	<hollisb>	aliguori: I am behind on my changelogs, so it's certainly possible :)
14:00	<aliguori>	yup
14:00	<aliguori>	a few changesets starting at 5b433b4fca34e8a9a3c2eb932ffa0e2ae8594e94
14:00	<aliguori>	tonfa, the newest user manual has docs on how to use
14:01	<tonfa>	hollisb: but why having more than one guest ?
14:01	<hollisb>	tonfa: hm?
14:01	<tonfa>	action->guest is an array
14:02	<secureboot>	hollisb: i just don't understand what happens on interrupt - i thought it was delivered to guests
14:02	<hollisb>	secureboot: dom0 is also a "guest"
14:02	<secureboot>	hollisb: you seem to be saying it goes to dom0... perhaps via backend driver?
14:03	<secureboot>	so interrupt goes to dom0, which has a handler. Then what?
14:03	<hollisb>	secureboot: you might want to look into this a bit more first
14:03	<tonfa>	hollisb: in irq.c __do_IRQ_guest, it iterates through action->guest. I don't understand why it could be useful to send the irq to more than one guest.
14:04	<secureboot>	hollisb: okay, where?
14:04	<hollisb>	a good rule of thumb is figure out the easy answers yourself, and only ask for help on the hard ones :)
14:04	<secureboot>	hollisb: i can't seem ti find a cogent discussion.
14:04	<hollisb>	secureboot: then you may have to turn to the code
14:05	<hollisb>	tonfa: I don't know why
14:05	<tonfa>	ok, btw i like the idea of having drivers in separate domains
14:06	<tonfa>	it could be used as a kind of sandboxing
14:06	<hollisb>	exactly
14:06	<tonfa>	something like nooks maybe
14:06	<tonfa>	but with probably more overhead
14:06	<hollisb>	unfortunately you can't implement effective isolation on legacy x86 systems
14:07	<tonfa>	ok. like if the hardware crash, everything will freeze ?
14:08	<hollisb>	that too. but even a bug in the device driver in a driver domain could bring down the hypervisor (and thus the whole machine)
14:08	<hollisb>	Intel and AMD have announced that in the future they will add hardware to fix that problem
14:08	<hollisb>	other companies (e.g. IBM) already have such hardware
14:09	<tonfa>	any pointer to the reasons the hypervisor will crash ?
14:09	<secureboot>	tonfa: DMA can't write to any/read from any memory, right?
14:09	<f>	hollisb: to be exact the hardware will be in the chipsets, so it depends on the chipset vendor
14:09	<hollisb>	secureboot: DMA can, that's exactly the problem
14:10	<tonfa>	ok, i think i understood
14:10	<hollisb>	f: hmm, I think AMD said theirs would be on the processor
14:10	<secureboot>	tonfa: this is actually documented outside of code, fortunately:)
14:10	<f>	hollisb: dma protection yes, but not the full IOMMU spec they recently published
14:10	<hollisb>	hmm ok
14:11	<secureboot>	hollisb: i'm still a little hung up on this - to understand even generally how device sharing between domUs works, i have to read the code? there's no high-level overview at all that describes how data safety/separationi in devices in assured?
14:11	<f>	and with intel it's the same of course
14:11	<tonfa>	secureboot: front end / backend drivers
14:11	<hollisb>	secureboot: there may be, but I don't have the time to look for it
14:11	<hollisb>	secureboot: and I wouldn't count on it existing
14:11	<secureboot>	hollisb: okay, thanks.
14:12	<secureboot>	tonfa: yeah - but how does the backend driver multiplex the data? how does it determine what to send to which domu?
14:12	<sdague>	secureboot: all the guest domains have virtual device drivers
14:12	<secureboot>	sdague: okay, yeah
14:12	<sdague>	secureboot: they are all point to point connections
14:12	<secureboot>	sdague: to a backend?
14:12	\|-\|	jerone [~jerone@adsl-71-145-138-37.dsl.austtx.sbcglobal.net] has quit [Quit: Leaving]
14:12	<sdague>	notice how you get a new vif interface in dom0 when you add a nic to a domU
14:13	<secureboot>	sdague: yeah
14:13	<secureboot>	sdague: and vbds
14:13	<aliguori>	tonfa, every guest (even non-privileged guests) get NMIs and event channel interrupts
14:13	<tonfa>	ok
14:13	<sdague>	secureboot: same basic idea, there is a point to point connection for the device in dom0
14:14	<hollisb>	aliguori: we are talking about __do_IRQ_guest in particular though
14:14	<hollisb>	aliguori: which is only for real device IRQs AFAICS
14:14	<tonfa>	nmi are handled by entry.S, no ?
14:15	<tonfa>	(on x86)
14:15	<aliguori>	hollisb, ah, yes, that's just for delivering pirqs
14:16	<secureboot>	sdague: so if you have 6 domUs which want a access a device, you'll have 6 virtual devices in the dom0? Given an IRQ, how does the dom0 know which virtual device the IRQ should go to then?
14:16	<aliguori>	in principle though, driver domains should receive their own pirqs though
14:16	<sdague>	secureboot: give an example device
14:16	<aliguori>	i don't know if that's how it works, but i presume so
14:16	<secureboot>	sdague: network card
14:17	<sdague>	the domUs don't get access to the network card directly
14:17	<sdague>	they get vifs
14:17	<aliguori>	secureboot, event channel notifications are a type of IRQ
14:17	<secureboot>	sdague: sure - but when an IRQ for the network card comes into dom0, how does it know which VIF should get the data?
14:17	<sdague>	then you setup up bridging or routing between the vifs and the real network cards in dom0
14:17	<sdague>	through the normal ways that linux can act as a router
14:18	<aliguori>	secureboot, through xenstore/xend magic, an event channel and shared memory is allocated between dom0 and domU and that is used for those purposes
14:20	<secureboot>	okay - this makes sense, i think. An interrupt occurs on the network card, Linux figures out which vif it should go to, and then Xen signals on the event channel to (what - i'm assuming front-end driver), which gets the data from the page the domU gave dom0 to return the data on
14:21	<aliguori>	secureboot, yup
14:21	<aliguori>	secureboot, it's a bit more complex than that b/c the network drivers do page flipping
14:21	<aliguori>	but that's the general idea
14:21	<secureboot>	aliguori: okay, thanks
14:21	<secureboot>	then what is http://wiki.xensource.com/xenwiki/XenIntro#head-cbf18c0f662adf210024bf449d3329c509be4fa0 talking about?
14:22	<aliguori>	i've not read this and I don't know who this guy is :-)
14:22	<secureboot>	basically, what does do_IRQ_guest do then?
14:22	<secureboot>	it seems to pass one interrupt to a lot of guests that have registered on it, according to this
14:23	<aliguori>	i've not read that text so i cannot confirm/deny what it says
14:24	<secureboot>	i guess i'm just a bit confused as to how multiple guests can be registered on an IRQ
14:24	<secureboot>	that's always been my fundamental question, i think.
14:27	<secureboot>	or maybe this isn't possible...
14:27	<secureboot>	if it wasn't possible, i don't think i'd have any questions at all :)
14:28	<secureboot>	EVTCHNOP_bind_pirq description seems to indicate that it may not be...
14:28	<aliguori>	two guests cannot bind to the same device. i don't know if it's possible for two guests to share a single pirq
14:29	<aliguori>	b/c it's not uncommon for multiple devices to use the same pirq
14:29	<aliguori>	and the interrupt handler has to figure out whether the interrupt was meant for it
14:29	<secureboot>	aliguori: which was my question exactly
14:29	<aliguori>	i would imagine that you could not do that with guests though
14:29	<aliguori>	b/c one guest could lie and say it was meant for it, and therefore starve another guest of it's devices irqs
14:30	<secureboot>	aliguori: i didn't see how that would happen - given an interrupt on a pirq, if there are multiple guests on it, how do you tell who it was for? and how to do mediate the data on the device?
14:30	<secureboot>	if that's not supported, i don't think there was ever anything i didn't get... i think i just got ahold of a bad description of pirqs in Xen, maybe.
14:30	<aliguori>	yes
14:30	<aliguori>	i think so too
14:31	<secureboot>	aliguori: well, thanks for clearing that up - I appreciate the patience.
14:31	<aliguori>	np
14:31	<secureboot>	time to start working with the code then.
14:36	\|-\|	ivan_ [~ikelly@83-71-74-95.b-ras1.lmk.limerick.eircom.net] has joined #xen
14:38	\|-\|	ivan [~ikelly@83-71-74-95.b-ras1.lmk.limerick.eircom.net] has quit [Ping timeout: 480 seconds]
14:40	\|-\|	aliguori [~anthony@cpe-70-116-13-229.austin.res.rr.com] has quit [Quit: Ex-Chat]
14:40	\|-\|	mdday [~mdday@bi01p1.nc.us.ibm.com] has quit [Quit: mdday]
14:48	\|-\|	mdday [~mdday@bi01p1.nc.us.ibm.com] has joined #xen
14:49	\|-\|	mdday [~mdday@bi01p1.nc.us.ibm.com] has quit [Quit: ]
14:54	\|-\|	mdday [~mdday@bi01p1.nc.us.ibm.com] has joined #xen
14:58	\|-\|	jerone [~jerone@adsl-71-145-138-37.dsl.austtx.sbcglobal.net] has joined #xen
15:03	\|-\|	cableman [~cableman@3E6B4910.rev.stofanet.dk] has quit [Remote host closed the connection]
15:04	\|-\|	ns [~niv@129.33.1.37] has joined #xen
15:08	\|-\|	mdday [~mdday@bi01p1.nc.us.ibm.com] has quit [Quit: mdday]
15:18	\|-\|	rusty [~rusty@bh02i525f01.au.ibm.com] has joined #xen
15:19	\|-\|	mdday [~mdday@bi01p1.nc.us.ibm.com] has joined #xen
15:31	\|-\|	anball [~anball@bi01p1.nc.us.ibm.com] has quit [Quit: Leaving]
15:33	\|-\|	aliguori [~anthony@w-mob201-128-62-101-251.public.utexas.edu] has joined #xen
16:01	\|-\|	aliguori [~anthony@w-mob201-128-62-101-251.public.utexas.edu] has quit [Quit: Ex-Chat]
16:03	\|-\|	bestorga [~bestorga@129.33.1.37] has joined #xen
16:28	\|-\|	secureboot [~secureboo@coriakin.cse.psu.edu] has quit [Quit: secureboot]
16:43	\|-\|	riel changed nick to unriel
17:33	\|-\|	bestorga [~bestorga@129.33.1.37] has quit [Quit: Leaving]
17:35	\|-\|	mdday [~mdday@bi01p1.nc.us.ibm.com] has quit [Quit: mdday]
17:43	\|-\|	zwane [~zwane@S0106001217d33fa6.vc.shawcable.net] has joined #xen
17:44	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has quit [Ping timeout: 480 seconds]
17:48	\|-\|	hollisb [~hollisb@pixpat.austin.ibm.com] has quit [Quit: leaving]
17:53	\|-\|	ns [~niv@129.33.1.37] has quit [Quit: Leaving]
17:59	\|-\|	ns [~niv@129.33.1.37] has joined #xen
18:01	\|-\|	dansmith_ [~dan@129.33.1.37] has quit [Quit: Leaving]
18:03	\|-\|	ns [~niv@129.33.1.37] has left #xen []
18:12	\|-\|	aliguori [~anthony@cpe-70-116-13-229.austin.res.rr.com] has joined #xen
18:15	\|-\|	horms [~horms@vagw.valinux.co.jp] has joined #xen
18:32	\|-\|	rharper [~rharper@pixpat.austin.ibm.com] has quit [Quit: Leaving]
18:48	\|-\|	dansmith [~dan@c-24-21-32-166.hsd1.or.comcast.net] has joined #xen
19:07	\|-\|	Edgan [~edgan@64-42-67-14.atgi.net] has quit [Quit: Client exiting]
19:19	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
19:28	\|-\|	stekloff [~stekloff@129.33.1.37] has quit [Quit: Leaving]
19:29	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has joined #xen
19:31	\|-\|	Shoragan [~shoragan@d072.apm.etc.tu-bs.de] has quit [Quit: ]
19:34	\|-\|	dansmith [~dan@c-24-21-32-166.hsd1.or.comcast.net] has quit [Quit: Leaving]
19:35	\|-\|	ivan_ [~ikelly@83-71-74-95.b-ras1.lmk.limerick.eircom.net] has quit [Quit: leaving]
19:43	\|-\|	Basic_py [~Basic@mailserver.dayport.com] has quit [Quit: Leaving]
19:46	\|-\|	ns [~nivedita@c-67-171-181-157.hsd1.or.comcast.net] has joined #xen
19:57	\|-\|	Edgan [~Edgan@71.83.124.204] has joined #xen
19:57	<Edgan>	f: I tried kpartx. It seems to do exactly what I need.
20:08	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has quit [Quit: mdday]
20:15	\|-\|	bestorga [~bestorga@c-24-20-32-112.hsd1.or.comcast.net] has joined #xen
20:17	\|-\|	johnlev [~johnlev@nwkea-socks-1.sun.com] has quit [Ping timeout: 480 seconds]
20:25	\|-\|	bestorga [~bestorga@c-24-20-32-112.hsd1.or.comcast.net] has quit [Ping timeout: 480 seconds]
20:28	\|-\|	gerrit [~gerrit@129.33.1.37] has quit [Ping timeout: 480 seconds]
20:51	\|-\|	tab [~tab@darwin.snarc.org] has joined #xen
20:51	\|-\|	tab_ [~tab@darwin.snarc.org] has quit [Read error: Connection reset by peer]
21:12	\|-\|	f [~f@p54899015.dip0.t-ipconnect.de] has quit [Ping timeout: 480 seconds]
21:15	\|-\|	hikenboot [~hikenboot@c-24-218-84-234.hsd1.ma.comcast.net] has joined #xen
21:16	<hikenboot>	hello ...I have found a place that provides CentOS domU images but they are small. I wish to expand the image to take 6 gig of space instead of 1gig so i can install a lot of software into it..is there an easy way to expand image files?
21:19	\|-\|	horms [~horms@vagw.valinux.co.jp] has quit [Quit: Leaving]
21:19	\|-\|	horms [~horms@vagw.valinux.co.jp] has joined #xen
21:20	\|-\|	michal` [~michal@www.rsbac.org] has quit [Ping timeout: 480 seconds]
21:24	<Edgan>	hikenboot: Should be possible with dd using an offset, and resizing the filesystem.
21:25	<Edgan>	hikenboot: They used ext3, right?
21:25	\|-\|	michal` [~michal@www.rsbac.org] has joined #xen
21:31	<Edgan>	hikenboot: working on it now
21:35	<Edgan>	hikenboot: dd if=/dev/zero of=test.img seek=1024 bs=1M count=5120
21:35	<Edgan>	hikenboot: That would add 5gb of space to a file that is exactly 1gb in size, making it 6gb.
21:35	<Edgan>	hikenboot: then e2fsck -f test.img ; resize2fs test.img
21:35	<Edgan>	hikenboot: Will fsck and resize it for you
21:37	<Edgan>	hikenboot: Then mount test.img /mnt/test should work and show you a 6gb filesystem.
21:52	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has joined #xen
21:58	\|-\|	hollisb [~hollisb@user-0vvde2g.cable.mindspring.com] has joined #xen
22:13	\|-\|	mdday [~mdday@cpe-024-163-120-222.nc.res.rr.com] has quit [Quit: mdday]
22:32	\|-\|	Suit_ [~SCOwillgo@cl-221.ams-05.nl.sixxs.net] has joined #xen
22:32	\|-\|	Suit [~SCOwillgo@cl-221.ams-05.nl.sixxs.net] has quit [Read error: Connection reset by peer]
22:38	<hikenboot>	Edgan thanks for the info i actually rsynced two different loop files one 1 gig the other 6 gig and deleted the smaller but that would work better!
22:38	<Edgan>	hikenboot: :)
22:38	<hikenboot>	good night
22:38	\|-\|	hikenboot [~hikenboot@c-24-218-84-234.hsd1.ma.comcast.net] has quit [Quit: Leaving]
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has left #xen [Rotating Logs]
22:59	\|-\|	VS_ChanLog [~stats@ns.theshore.net] has joined #xen
23:14	\|-\|	Basic_py [~Basic@warden.real-time.com] has joined #xen
23:15	\|-\|	f [~f@p54898F37.dip0.t-ipconnect.de] has joined #xen
23:16	\|-\|	hollisb [~hollisb@user-0vvde2g.cable.mindspring.com] has quit [Quit: leaving]
---	Log	closed Tue Feb 28 00:00:42 2006