Hi, all.

I ran into the following problem. One of the zfs data disk is become unavailable and solaris marked it as bad for a too long time.

As I understand, disk must be mark offline if disk timeout grater than 10 seconds  (scsi_timeout) * scsi_retry == 30 seconds, but in log I see that I/O was degraded more than 5 minutes! How can I fix It?

CONFIGURATION

scsi_vhci.conf

# File is managed by Ansible

name="scsi_vhci" class="root";

load-balance="round-robin";

auto-failback="enable";

ddi-forceload =

"misc/scsi_vhci/scsi_vhci_f_asym_sun",

"misc/scsi_vhci/scsi_vhci_f_asym_lsi",