Re: [Libvirt-cim] [PATCH 1 of 2] ELEC: seg faults by wrong client given object path

Dan Smith wrote:

HE> + s = get_elec_by_name(broker, reference, name, _inst); HE> + if (s.rc != CMPI_RC_OK) HE> + goto out; HE> + HE> + s = cu_validate_ref(broker, reference, *_inst); HE> +

Assuming the caller checks the status, there shouldn't be an issue, but maybe we shouldn't set _inst until we've validated the path?

Yes, you are absolutely right. I've changed this in the patches for ELEC, RASD and VSSD. -- Regards Heidi Eckhart Software Engineer IBM Linux Technology Center - Open Hypervisor