It looks like it completed the COPY statement for the table that it was backing up right before the error message occurred. I also do not see anything in the backup file that indicates what is happening.
Is the entire process occurring on the same host, or is across a network? If the latter, could it be due to timeouts from ssh, or tunnels, or internal load balancers/connection poolers?