I'm hoping someone has run into this problem.
i'm using 2 aoc-sat2-mv8 controllers with 15 identical hitachi ultrastar 750GB drives running 8.2-STABLE zfs 5/28 for the pools.
the problem is if i create a raid6 vdev across these controllers the system instantly reboots if i try writing to the vdev. if i create a mirror across the controllers, i can run a DD type file creation for a minute or so then the system crashes.
i can create a cross controller raid0 with all 15 drives and it works fine.
i can create raidz2 vdevs on each controller and these work fine. if i use the exact same drives that just worked fine in a nested raidz2 (raid60) again, instant crash.
i have a similar setup in another server using WD RE4 1TB drives working fine cross controllers, same aoc-sat2-mv8 controller type and that works great.
any ideas how i can track down the cause of this?
i'm using 2 aoc-sat2-mv8 controllers with 15 identical hitachi ultrastar 750GB drives running 8.2-STABLE zfs 5/28 for the pools.
the problem is if i create a raid6 vdev across these controllers the system instantly reboots if i try writing to the vdev. if i create a mirror across the controllers, i can run a DD type file creation for a minute or so then the system crashes.
i can create a cross controller raid0 with all 15 drives and it works fine.
i can create raidz2 vdevs on each controller and these work fine. if i use the exact same drives that just worked fine in a nested raidz2 (raid60) again, instant crash.
i have a similar setup in another server using WD RE4 1TB drives working fine cross controllers, same aoc-sat2-mv8 controller type and that works great.
any ideas how i can track down the cause of this?