[zfs-discuss] ZFS SATA Hotswap Autoreplace

2007-10-22 Thread Jeff Creek
I am trying to test a new setup of NV74.  I have set up the system with ZFS 
boot.  Everything works fine until I pull a drive.  The system locks up when I 
try to run any command e.g. zpool status.  There is no communication with the 
server from then on.  It does not even respond to pings.

Setup -

Supermicro SC836TQ-R800 Chassis
Supermicro X7DB8 Motherboard
Supermicro SAT2-MV8 SATA controller
Hitatchi Deskstar Harddrives 2 80GB Mirrored and 4 1 TB RaidZ
Autoreplace set to ON

If i push the reset button the system will come back up ok.  I can then run 
zpool status and find which drive is 'bad'.  I would rather the system keep 
functioning so I can just run the zpool status and then replace the 'bad' drive.

Is this setup not supported?  Is there something I missed?

Thanks,

Jeff
 
 
This message posted from opensolaris.org
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss


Re: [zfs-discuss] ZFS SATA Hotswap Autoreplace

2007-10-22 Thread Eric Schrock
I don't think that SATA controller is supported by the SATA framework
(you may want to ask storage-discuss), so it's operating in legacy
(PATA) mode.  Legacy mode doesn't support hotplug.  The FMA code should
eventually diagnose this drive as faulty, and there is a fix in the
works to sidestep this roundtrip discussion with fmd for pathologically
broken drives.

- Eric

On Mon, Oct 22, 2007 at 09:16:53AM -0700, Jeff Creek wrote:
 I am trying to test a new setup of NV74.  I have set up the system with ZFS 
 boot.  Everything works fine until I pull a drive.  The system locks up when 
 I try to run any command e.g. zpool status.  There is no communication with 
 the server from then on.  It does not even respond to pings.
 
 Setup -
 
 Supermicro SC836TQ-R800 Chassis
 Supermicro X7DB8 Motherboard
 Supermicro SAT2-MV8 SATA controller
 Hitatchi Deskstar Harddrives 2 80GB Mirrored and 4 1 TB RaidZ
 Autoreplace set to ON
 
 If i push the reset button the system will come back up ok.  I can then run 
 zpool status and find which drive is 'bad'.  I would rather the system keep 
 functioning so I can just run the zpool status and then replace the 'bad' 
 drive.
 
 Is this setup not supported?  Is there something I missed?
 
 Thanks,
 
 Jeff
  
  
 This message posted from opensolaris.org
 ___
 zfs-discuss mailing list
 zfs-discuss@opensolaris.org
 http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

--
Eric Schrock, Solaris Kernel Development   http://blogs.sun.com/eschrock
___
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss