Hi,
I"ve been running simulations on our RAD 1 box. I pulled out a drive
(they're hot-swappable) and it starts reconstruction on the spare disk but
after a while the machine hangs with scsi bus timeouts and resets.
I have done the same test without the spare disk. LEt it run in degraded
mode for a bit, then manually raidhotadd-ing the spare disk partitions,
one by one, waiting for each slice to finish rebuilding before I add the
next, that seemed to work without a problem.
Has this happened to anyone?
Also, is there a way I can be notified if one of the disks goes down or
will I have to write something that monitors /proc/mdstatus?
Thank you in advance for any input.
Tamas.
