Re: [Xen-cim] [PATCH] Xen_OperatingSystem_setInstanceFromResource patch

[Gareth S Bestor <bestorga@xxxxxxxxxx>]

>...
>The assumption that property OSType == Linux will have to be fixed as well.

Yup. We may in fact need to consider dropping this class altogether and relying on in-band instrumentation (either
thru an inband cimom, or via a remoteCMPI provider), since moving forward with HVM guest's we can make little
or no assumptions about what binary code the guest happens to be running. As it was, always assuming "Linux" for para-virt was a bit of a stretch to begin with...

Note - there's no OS instance prescribed by the DMTF SV model, though it may show up when we extend the model to cover containers. So it is
really only there today as a convenient placeholder for virt unaware mgmt clients that expect to see an OS having of the CS. Whereas virt aware clients
should probably know that they'll need to look elsewhere (ie inband) to discover OS-related info about hypervisor guests.

I'm fine with leaving this hack in and patching it (with duct tape) or removing altogether.

- Gareth

Dr. Gareth S. Bestor
IBM Linux Technology Center
M/S DES2-01
15300 SW Koll Parkway, Beaverton, OR 97006
503-578-3186, T/L 775-3186, Fax 503-578-3186

Jim Fehlig <jfehlig@xxxxxxxxxx>

Jim Fehlig <jfehlig@xxxxxxxxxx>
Sent by: xen-cim-bounces@xxxxxxxxxxxxxxxxxxx

12/19/06 08:28 AM

To	"Subrahmanian, Raj" <raj.subrahmanian@xxxxxxxxxx>
cc	xen-cim@xxxxxxxxxxxxxxxxxxx
Subject	Re: [Xen-cim] [PATCH] Xen_OperatingSystem_setInstanceFromResource patch

Subrahmanian, Raj wrote:
> After the new updates to compile witth Xen-3.0.4, I found that we still
> could not enumerate Xen_OperatingSystem. The issue is that the pv_kernel
> parameter is null. This patch does a sanity check to ensure that the
> pv_kernel value is clean before it does anything. Also, it will not
> fault if the pv_kernel path does not look like it expects it to be.
> pv_kernel should not have been null, ut this initial patch will ensure
> that the CIMOM does not croak on bad returns from libXen.
>
> With this patch, I can enumerate Xen_OperatingSystem classes.
>
> Raj
>   

Need the patch :-).  The previous one is against pre-3.0.4.

I'm assuming you took care of the case where pv_kernel is null - which 
will always be the case if bootloader is used or guest is HVM.  We'll 
have to think about how to deal with the Xen_OperatingSystem class in 
these cases.  The assumption that property OSType == Linux will have to 
be fixed as well.

Jim


_______________________________________________
Xen-cim mailing list
Xen-cim@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-cim

_______________________________________________
Xen-cim mailing list
Xen-cim@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-cim