libvirt users - Mar 2012 - 转发: Error when executing virsh command to ESX

To Whom It May concern:
I found that the following error is caused by adding a scsi disk to install the
os, I remove the scsi disk and the error remove but still report 'out of
memory':
virsh # dumpxml win2003_122
internal error Invalid or not yet handled value '/vmfs/devices/genscsi/mp
x.vmhba0:C0:T1:L0' for VMX entry 'scsi0:1.fileName'


virsh # list --all
 Id Name                 State
----------------------------------
 80 win2003_122          running
  - winxp121             shut off

virsh # nodeinfo
error: failed to get node information
error: out of memory

virsh # dumpxml win2003_122
error: out of memory




li_fuqiong

???? li_fuqiong
????? 2012-03-21 15:24
???? libvirt-users
??? Error when executing virsh command to ESX
To Whom It May Concern:

I use virsh windows 0.9.3 to connect an esx server 4.0 with no_verify=1. The Esx
server is intel 2 x 3.19 GHZ CPU, 8G memroy and 2 TB hard disk.
following is my execute commands and results.

virsh -c esx://xxx.xxx.xxx.x?no_verify=1

virsh # list all

 Id Name                 State
----------------------------------
 80 win2003_122          running

virsh # nodeinfo
error: failed to get node information
error: out of memory

virsh # dumpxml win2003_122
error: internal error Invalid or not yet handled value
'/vmfs/devices/genscsi/mp
x.vmhba0:C0:T1:L0' for VMX entry 'scsi0:1.fileName'

virsh # edit win2003_122
error: internal error Invalid or not yet handled value
'/vmfs/devices/genscsi/mp
x.vmhba0:C0:T1:L0' for VMX entry 'scsi0:1.fileName'

virsh # destroy win2003_122
Domain win2003_122 destroyed

virsh # start win2003_122
Domain win2003_122 started

virsh # edit win2003_122
error: internal error Invalid or not yet handled value
'/vmfs/devices/genscsi/mp
x.vmhba0:C0:T1:L0' for VMX entry 'scsi0:1.fileName'

question:
1.Why error when executeing dumpxml and edit command? How to fix it?
2.Why out of memory when nodeinfo. How to fix it? I have only one test vm on my
esx server.


Thanks in advance.





Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://listman.redhat.com/archives/libvirt-users/attachments/20120321/3e9a0b20/attachment.htm>

On 2012?03?21? 15:49, li_fuqiong wrote:
> To Whom It May concern:
> I found that the following error is caused by adding a scsi disk to
> install the os, I remove the scsi disk and the error remove but still
> report 'out of memory':
When trying to ask for help, it's better to tell which hypervisor
driver you are using.
> virsh # dumpxml win2003_122
> internal error Invalid or not yet handled value
'/vmfs/devices/genscsi/mp
> x.vmhba0:C0:T1:L0' for VMX entry 'scsi0:1.fileName'
 From the source and per the error you got, we can known you are
using VMWare driver though.

VMWare driver parses the disk conf from .vmx file, and it recongnizes
the disk type with the disk file name suffix. Think you will fix
the problem by append ".vmdk" to the scsi disk you use in .vmx.
> virsh # list --all
> Id Name State
> ----------------------------------
> 80 win2003_122 running
> - winxp121 shut off
> virsh # nodeinfo
> error: failed to get node information
> error: out of memory
VMWare driver doesn't support nodeinfo, it's expected to see error
lile "this function is not supported by the connection driver". While
you see this, it's a bug. Could you try with libvirt 0.9.10 to see
if the problem is fixed yet?

Regards,
Osier