Controller failover and replacement on the StorEdge 3310

AUTOMATIC FAILOVER

If one controller fails in a redundant controller configuration, the surviving controller takes over for the failed controller until it is replaced. The surviving controller then manages all processes for the failed controller. The surviving controller is always the primary controller, regardless of its original status, and any replacement controller afterwards will become the secondary controller.

Failover and failback processes are transparent to the host.

Controllers are hot-swappable.

MANUAL FAILOVER

The following output is with two luns mapped to the primary controller on channel 3 with one host connection to channel 3:

# devfsadm -C

# format

Searching for disks...done

c3t0d1: configured with capacity of 998.00MB

AVAILABLE DISK SELECTIONS:

0. c0t0d0 <ST39140A cyl 17660 alt 2 hd 16 sec 63>

/pci@1f,0/pci@1,1/ide@3/dad@0,0

1. c1t3d0 <SUN4.2G cyl 3880 alt 2 hd 16 sec 135>

/pci@1f,0/pci@1/pci@1/SUNW,isptwo@4/sd@3,0

2. c3t0d0 <SUN-StorEdge2310-0325 cyl 51712 alt 2 hd 64 sec 64> <===== 3310 lun

/pci@1f,0/pci@1/pci@4/scsi@5/sd@0,0

3. c3t0d1 <SUN-StorEdge2310-0325 cyl 998 alt 2 hd 64 sec 32> <===== 3310 lun

/pci@1f,0/pci@1/pci@4/scsi@5/sd@0,1

Specify disk (enter its number): ^D

In this example I put a filesystem on disk 3:c3t0d1, mounted it on /mnt and copied /etc to it. Then I reassigned the lun on the array from the primary controller to the secondary controller:

# format

Searching for disks...done

AVAILABLE DISK SELECTIONS:

0. c0t0d0 <ST39140A cyl 17660 alt 2 hd 16 sec 63>

/pci@1f,0/pci@1,1/ide@3/dad@0,0

1. c1t3d0 <SUN4.2G cyl 3880 alt 2 hd 16 sec 135>

/pci@1f,0/pci@1/pci@1/SUNW,isptwo@4/sd@3,0

2. c3t0d0 <SUN-StorEdge2310-0325 cyl 51712 alt 2 hd 64 sec 64>

/pci@1f,0/pci@1/pci@4/scsi@5/sd@0,0

3. c3t0d1 <drive type unknown> <====== see the "drive type inknown"

/pci@1f,0/pci@1/pci@4/scsi@5/sd@0,1

4. c3t1d0 <SUN-StorEdge2310-0325 cyl 998 alt 2 hd 64 sec 32> <====== this is the new location of the same lun I just moved

/pci@1f,0/pci@1/pci@4/scsi@5/sd@1,0

Specify disk (enter its number):

It now shows up as a "ghost" lun on disk 3, as well as a new disk 4:c3t1d0. For now the data is still available on the /mnt:

# cd /mnt

# ls (shows data)

# df -k

Filesystem kbytes used avail capacity Mounted on

/dev/dsk/c0t0d0s0 7649269 1567985 6004792 21% /

/proc 0 0 0 0% /proc

fd 0 0 0 0% /dev/fd

mnttab 0 0 0 0% /etc/mnttab

swap 842184 16 842168 1% /var/run

swap 842496 328 842168 1% /tmp

/dev/dsk/c3t0d1s0 30799 2399 25321 9% /mnt <======see the old device entry: c3t0d1s0

Then I unmounted /mnt and was able to mount with the new device path and the data was still there:

# umount /mnt

# df -k

Filesystem kbytes used avail capacity Mounted on

/dev/dsk/c0t0d0s0 7649269 1567985 6004792 21% /

/proc 0 0 0 0% /proc

fd 0 0 0 0% /dev/fd

mnttab 0 0 0 0% /etc/mnttab

swap 842104 16 842088 1% /var/run

swap 842416 328 842088 1% /tmp

# mount /dev/dsk/c3t1d0s0 /mnt

# df -k

Filesystem kbytes used avail capacity Mounted on

/dev/dsk/c0t0d0s0 7649269 1567985 6004792 21% /

/proc 0 0 0 0% /proc

fd 0 0 0 0% /dev/fd

mnttab 0 0 0 0% /etc/mnttab

swap 842088 16 842072 1% /var/run

swap 842400 328 842072 1% /tmp

/dev/dsk/c3t1d0s0 30799 2399 25321 9% /mnt <====== new device mounted: c3t1d0s0

# cd /mnt

# ls (data still available)

After a reboot, format was normal again and I was able to mount the lun and still have data intact. So, the manual failover worked and the data remained available the whole time but you will need to be sure the mounted device is updated in vfstab for the next reboot.

Daniel Martinez

d.m@sun.com