On 01/21/13 20:04, Eric Blake wrote:
On 01/21/2013 03:53 AM, Peter Krempa wrote:
>> I'm slightly against this patch as I prefer a crashing daemon to silent
>> propagation of bug.
>
> Hm, yeah. I agree. The snapshot metaroot patch that caused this problem
> expects that all snapshots except the metaroot have parents so this may
> disclose other potential bugs.
Hmm, I wonder if the real bug is that we somehow earlier managed to
create a snapshot without setting the metaroot as its initial parent.
Yes it was an earlier error where the snapshot did not have a parent
filled out as when the saving of the definition XML failed. The code
path there printed a warning and did not fill the parent. See patch 3/4.
Peter