[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-users] virtualization doesn't work with xen 4.4



On Sun, Apr 27, 2014 at 08:15:08PM +0200, Jacek Konieczny wrote:
> On 2014-04-27 12:23, Michael Mair-Keimberger wrote:
> > I've recently updated my xen setup from 4.1 to 4.4. Now it's impossible
> > for me to start any (basic) vm.
> 
> Any basic HVM, I guess. I don't think the problem appears on PV virtual
> machines.

Honestly i havn't checked others than hvm but i guess so :)

> 
> > My basic config looks like this atm:
> > 
> >> on_poweroff='destroy'
> >> on_reboot='destroy'
> >> on_crash='destroy'
> >>
> >> disk=['file:/home/michael/xen/win7x64s.img,hda,w','file:/home/michael/xen/dvd/win7x64.iso,hdc:cdrom,r']
> >> vif=['mac=DE:AD:CA:5E:13:37,bridge=br0,type=ioemu']
> >>
> >> boot='d'
> >>
> >> name = 'HVM_domU'
> >> builder = 'hvm'
> >> memory = 256
> >> vcpus = 2
> >> vnc = 1
> >> vnclisten = '0.0.0.0'
> >> vncdisplay = 1
> > 
> > When i try to start this, xen immediately kills the vm again. 
> > From the logs i get following:
> 
> [...]
> 
> > xl dmesg only shows one returning error, which didn't help me at all as i
> > couldn't find any usefull help about this message:
> > 
> >> (XEN) io.c:204:d2 MMIO emulation failed @ 0008:ffff1f50: 4a 00 00 00 00 00 
> >> 00 00 36 99
> 
> I have already reported a very simila problem, here:
> 
> http://lists.xen.org/archives/html/xen-users/2014-04/msg00084.html
> 
> For me this occurs only with HVM guests using the default 'qemu-xen'
> device model.
> 
> PV and PVH with 'qemu-xen-traditional' device model work for me.
> 
> I assumed the problem was related to the system qemu I use instead of
> the qemu built from the Xen sources (I prefer not to maintain multiple
> qemu builds for a distribution I work for). I am not sure if this would
> occur with the 'qemu-xen' from the Xen sources in my case.

Thx for pointing out your report. It looks indeed similiar. However, i
did some test and still couldn't get it to work at all. Following
configuration were tested:

xen build with bundled qemu and device_model="qemu-xen"
> libxl: detail: libxl_dom.c:195:numa_place_domain: NUMA placement candidate 
> with 1 nodes, 8 cpus and 7981 KB free selected
> xc: detail: elf_parse_binary: phdr: paddr=0x100000 memsz=0x9f6e4
> xc: detail: elf_parse_binary: memory: 0x100000 -> 0x19f6e4
> xc: detail: VIRTUAL MEMORY ARRANGEMENT:
> Loader:        0000000000100000->000000000019f6e4
> Modules:       0000000000000000->0000000000000000
> TOTAL:         0000000000000000->000000000f800000
> ENTRY ADDRESS: 0000000000100000
> xc: detail: PHYSICAL MEMORY ALLOCATION:
> 4KB PAGES: 0x0000000000000200
> 2MB PAGES: 0x000000000000007b
> 1GB PAGES: 0x0000000000000000
> xc: detail: elf_load_binary: phdr 0 at 0x7f9312c50000 -> 0x7f9312ce656d
              
xen build with bundled qemu and device_model="qemu-xen-traditional"
using device_model_override to point to system qemu (used qemu-2.0):
> libxl: detail: libxl_dom.c:195:numa_place_domain: NUMA placement candidate 
> with 1 nodes, 8 cpus and 7981 KB free selected
> xc: detail: elf_parse_binary: phdr: paddr=0x100000 memsz=0x9f6e4
> xc: detail: elf_parse_binary: memory: 0x100000 -> 0x19f6e4
> xc: detail: VIRTUAL MEMORY ARRANGEMENT:
> Loader:        0000000000100000->000000000019f6e4
> Modules:       0000000000000000->0000000000000000
> TOTAL:         0000000000000000->000000000fc00000
> ENTRY ADDRESS: 0000000000100000
> xc: detail: PHYSICAL MEMORY ALLOCATION:
> 4KB PAGES: 0x0000000000000200
> 2MB PAGES: 0x000000000000007d
> 1GB PAGES: 0x0000000000000000
> xc: detail: elf_load_binary: phdr 0 at 0x7f2a1d548000 -> 0x7f2a1d5de56d
> libxl: error: libxl_dm.c:1371:device_model_spawn_outcome: domain 37 device 
> model: spawn failed (rc=-3)
> libxl: error: libxl_create.c:1186:domcreate_devmodel_started: device model 
> did not start: -3
> libxl: error: libxl_dm.c:1475:kill_device_model: Device Model already exited

xen build with system qemu support (qemu-2.0) and device_model="qemu-xen"
> libxl: detail: libxl_dom.c:195:numa_place_domain: NUMA placement candidate 
> with 1 nodes, 8 cpus and 7981 KB free selected
> xc: detail: elf_parse_binary: phdr: paddr=0x100000 memsz=0x9f6e4
> xc: detail: elf_parse_binary: memory: 0x100000 -> 0x19f6e4
> xc: detail: VIRTUAL MEMORY ARRANGEMENT:
> Loader:        0000000000100000->000000000019f6e4
> Modules:       0000000000000000->0000000000000000
> TOTAL:         0000000000000000->000000000f800000
> ENTRY ADDRESS: 0000000000100000
> xc: detail: PHYSICAL MEMORY ALLOCATION:
> 4KB PAGES: 0x0000000000000200
> 2MB PAGES: 0x000000000000007b
> 1GB PAGES: 0x0000000000000000
> xc: detail: elf_load_binary: phdr 0 at 0x7f26ba560000 -> 0x7f26ba5f656d
> libxl: error: libxl_dm.c:1371:device_model_spawn_outcome: domain 23 device 
> model: spawn failed (rc=-3)
> libxl: error: libxl_create.c:1186:domcreate_devmodel_started: device model 
> did not start: -3
> libxl: error: libxl_dm.c:1475:kill_device_model: Device Model already exited

xen build with system qemu support (qemu-2.0) and
device_model="qemu-xen-traditional"
> libxl: detail: libxl_dom.c:195:numa_place_domain: NUMA placement candidate 
> with 1 nodes, 8 cpus and 7981 KB free selected
> xc: detail: elf_parse_binary: phdr: paddr=0x100000 memsz=0x9f6e4
> xc: detail: elf_parse_binary: memory: 0x100000 -> 0x19f6e4
> xc: detail: VIRTUAL MEMORY ARRANGEMENT:
> Loader:        0000000000100000->000000000019f6e4
> Modules:       0000000000000000->0000000000000000
> TOTAL:         0000000000000000->000000000f800000
> ENTRY ADDRESS: 0000000000100000
> xc: detail: PHYSICAL MEMORY ALLOCATION:
> 4KB PAGES: 0x0000000000000200
> 2MB PAGES: 0x000000000000007b
> 1GB PAGES: 0x0000000000000000
> xc: detail: elf_load_binary: phdr 0 at 0x7ff502be8000 -> 0x7ff502c7e56d


Any ideas? I guess i have to check xen with other than hvm guests?

thx

greetings
Michael


> 
> Greets,
>     Jacek
> 
> _______________________________________________
> Xen-users mailing list
> Xen-users@xxxxxxxxxxxxx
> http://lists.xen.org/xen-users

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxx
http://lists.xen.org/xen-users


 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.