When the rebuild cache is shared amongst multiple nodes, there can be collisions when more than one
node tries to stash into the cache.
This solution, as much as it works, is less than optimal. Basically, all nodes do their build and
the last one to store in the cache wins.
A better solution would be for nodes to announce their intention to build for a given hash and any
other nodes that will do the same wait for the first node to finish and store it. Subsequent nodes
then just use the first node's stored result.
should be landed to b1_8. This is not important until we have build nodes sharing their cache, which we don't have at the moment.