I had a similar issue with one of my zpools the other day, granted not to the extent of 6 drives. 3 of my drives kept returning checksum errors, pool was offlined multiple times, replaced disks, and replaced cables with the issue still occurring randomly and sporadically. I began noticing loads of virtual AHCI lane failures, to the point where it was impossible to shut the system down and if I was lucky enough to get it to boot, it would take upwards of an hour or more. Long story short, multiple SATA port failures, both with the marvel, as well as intel, controllers. RMA'd the MB, no issues since receiving a new one.