Spurious errors reported at startup of XDCR replication

Right after replication was started between two clusters, some of these messages appeared in the UI:
2012-12-17 07:22:06 - Error replicating vbucket 250: {badmatch, {error, all_nodes_failed, <<"Failed to grab remote bucket info from any of known nodes">>}}"
OR
2012-12-17 07:25:52 - Error replicating vbucket 470: {http_request_failed, "HEAD", "http://Administrator:*****@10.155.240.66:8092/UserInfo%2f470%3b7b7614ca1cceb5e1940bd6ef0b30d745/", {error, {error, {eaddrinuse, [

{lhttpc_client, send_request,1}

,

{lhttpc_client, execute,9}

,

{lhttpc_client, request,9}

]}}}}."

They did not continue, and did not seem to affect the data transfer.

Can they be resolved so that the customer doesn't see errors in an otherwise functioning replication?