Heidi Eckhart wrote:
# HG changeset patch # User Heidi Eckhart <heidieck@linux.vnet.ibm.com> # Date 1203507766 -3600 # Node ID c77500943eae17131c7d0522a99d16b3e2f43489 # Parent 60e38dd96ca6871856d64056305fc200d718b3a5 VSSD: validate client given object path
wbemgi 'http://localhost:5988/root/virt:KVM_VirtualSystemSettingData.InstanceID="Xen:kvm1-f8"' on a KVM only system is returning an instance, but should not.
wbemgi 'http://localhost:5988/root/virt:Xen_VirtualSystemSettingData.InstanceID="KVM:kvm1-f8"' on a KVM only system is seg faulting, but should not ;).
Signed-off-by: Heidi Eckhart <heidieck@linux.vnet.ibm.com>
diff -r 60e38dd96ca6 -r c77500943eae src/Virt_VSSD.c
- -static CMPIStatus enum_vssd(const CMPIObjectPath *reference, - const CMPIResult *results, - int names_only) + NAMESPACE(reference)); + + if (inst == NULL) { + cu_statusf(broker, s, + CMPI_RC_ERR_FAILED, + "Unable to init VirtualSystemSettingData instance"); + goto out; + } + + if (!instance_from_dom(dom, inst)) {
Probably not worth a resend, but since instance_from_dom() returns an int, this should be if (!instance_from_dom(dom, inst) != 1) {
+ + s = cu_validate_ref(broker, reference, *_inst);
If we encounter an error here, do we need to set a status message?
+ + free(name); + + out: + return s; -- Kaitlin Rupert IBM Linux Technology Center kaitlin@linux.vnet.ibm.com