When trying to replicate most or all jobs fail with "Error: 2014 - FATAL phys_cant_read_detail PHYS can't read /vmfs/volumes/vmname/vmname-flat.vmdk (Input/output error) [at phys_read:84]" This was showing using console replication.
原因
It was determined in this instance that the target datastore was running out of disk space, even though the error named the source disk.
解决办法
Enabling the replication option to "Check destination for free space" revealed that the target datastore was running low on disk space. Customer removed unneeded VMs on target datastore to free up space. Then replications were successful again.
