I have two VMWare ESXi 4.1 hosts, and am backing up the guests from a physical server over a 1 Gbps line. The guests are all on direct attached storage. A couple weeks ago the backups on one host began to crawl. For example, a 20 minute backup now takes 5 hours or so. This happens both with Veeam 6.1 and Backup Exec 2012. This problem only occurs for the guests on one host, whereas the other host is completely fine. Both servers use the same switches. The server that is slow to backup is able to host the VMs without any problem at all, and the vSphere client displays no errors or anything other than "normal" for the hardware status. Any idea whatsoever of what could be causing the problem?
