Kaitlin Rupert <kaitlin@linux.vnet.ibm.com> wrote on 2009-03-12 11:08:30:

> Just in case you missed it on the list..
>
> These messages you're seeing are expected failures.  The test case is
> trying to destroy a guest that no longer exists.  So you would expect to
> see errors in the provider debug, right?

 
 I debug the provider and get the error message. This failure because of
 the error message updates in VSMS provider. I send a patch with fix.

 Thanks for all your help =)

 Now, all the test failures are fixed for KVM. I will continue to cook up
 the new test cases to verify VSMS.RemoveResourceSettings(). On the next,
 I will start to fix failures for Xen/XenFV.

 
   
> --
> Kaitlin Rupert
> IBM Linux Technology Center
> kaitlin@linux.vnet.ibm.com
>
> ----- Message from Deepti B Kalakeri <deeptik@linux.vnet.ibm.com> on
> Mon, 09 Mar 2009 19:51:17 +0530 -----

>
> To:

>
> List for discussion and development of libvirt CIM <libvirt-cim@redhat.com>

>
> Subject:

>
> Re: [Libvirt-cim] Test Run Summary (Mar 09 2009): KVM on Fedora
> release 10 (Cambridge) with sfcb

>
>
>
> Guo Lian Yun wrote:
> >
> > libvirt-cim-bounces@redhat.com wrote on 2009-03-09 10:45:59:
> >
> > >
> > > libvirt-cim-bounces@redhat.com wrote on 2009-03-09 09:45:08:
> > >
> > > >
> > > >
> > > > =================================================
> > > > Test Run Summary (Mar 09 2009): KVM on Fedora release 10
> > > (Cambridge)with sfcb
> > > > =================================================
> > > > Distro: Fedora release 10 (Cambridge)
> > > > Kernel: 2.6.27.15-170.2.24.fc10.x86_64
> > > > libvirt: 0.4.5
> > > > Hypervisor: QEMU 0.9.1
> > > > CIMOM: sfcb sfcbd 1.3.3preview
> > > > Libvirt-cim revision: 829
> > > > Libvirt-cim changeset: 1aff0d0e9bf4
> > > > Cimtest revision: 642
> > > > Cimtest changeset: 087aaba26589
> > > > =================================================
> > > > FAIL      : 4
> > > > XFAIL     : 3
> > > > SKIP      : 5
> > > > PASS      : 134
> > > > -----------------
> > > > Total     : 146
> > > > =================================================
> > > > FAIL Test Summary:
> > > > ElementCapabilities - 03_forward_errs.py: FAIL
> > >   I've sent a patch with fix.
> > >
> > > > ElementConforms - 01_forward.py: FAIL
> > >   Same as above.
> > >
> > > > ElementConforms - 04_ectp_rev_errs.py: FAIL
> > >   Same as above.
> > >
> > > > VirtualSystemManagementService - 05_destroysystem_neg.py: FAIL
> >
> >     It seems this failure because that the provider has to free the
> > nonexistent domain in
> >     invoking method DestroySystem.
> >  
> >     Here is the debugger error in sfcb:
> >     libvir: Domain error : invalid domain pointer in virDomainFree
> >     libvir: QEMU error : Domain not found
> >     libvir: Domain error : invalid domain pointer in virDomainFree
> Daisy,
>
> The test case is written to verify the DestroySystem() returns an error
> when trying to remove non-existing domain
> and also when an invalid domain name is passed.
> The error statements returned seems to have changed.
>
> For ex:
> The err desc for DestroySystem() when non-existing domain is passed was
> previously
> CIM_ERR_FAILED:Unable to retrieve domain name
>
> is now changed to  CIM_ERR_NOT_FOUND: Unable to retrieve domain name:
> Error 0
>
> We need to change the way the error comparison is made.
> The comparison stmt like
>
> if err_desc.find(exp_value['desc']) >= 0: does not work, but
> if exp_value['desc'] in err_desc: works
>
> --
> Thanks and Regards,
> Deepti B. Kalakeri
> IBM Linux Technology Center
> deeptik@linux.vnet.ibm.com
>
> _______________________________________________
> Libvirt-cim mailing list
> Libvirt-cim@redhat.com
> https://www.redhat.com/mailman/listinfo/libvirt-cim