On Wed, Jul 21, 2021 at 2:29 PM Peter Krempa <pkrempa@xxxxxxxxxx> wrote:
On Wed, Jul 21, 2021 at 14:05:05 +0200, Kristina Hanicova wrote:
> We didn't always save status xml after generating new taint
> message, which resulted in it being deleted in case of a libvirtd
> restart. Some taint messages were preserved thanks to saving
> status xml separately at the end of the calling functions (which
> makes sense, because qemuDomainObjTaint was usually called there
> multiple times). But for special cases (e.g. When only few taint
> messages are generated) we need a separate function for
> generating them and saving status xml explicitly.
Saving the status XML is a very common operation which we in some cases
repeat a few times when doing an multi-step operation, thus we can
reasonably assume that saving the status XML in all cases when we are
adding a taint on a VM object is okay without the need to special case
operations which don't save the status XML as part of their code.
I didn't want to add saving status xml to the original function mainly because
of qemuDomainObjCheckTaint, where qemuDomainObjTaint could be potentially
called in multiple for cycles. I thought it would be ineffective, but I will
change the patch if that kind of situation is not likely.
of qemuDomainObjCheckTaint, where qemuDomainObjTaint could be potentially
called in multiple for cycles. I thought it would be ineffective, but I will
change the patch if that kind of situation is not likely.
Kristina
