Using Solstice DiskSuite 4.0 with the SPARCstorage Array 100

E

Solstice DiskSuite 4.0 is fully compatible with SPARCstorage Array 100 (SSA-100). In most situations, using SSA-100 disks with DiskSuite is just like using any other disks. There are, however, some characteristics of the SSA-100 of which DiskSuite users should be aware. The following sections detail these characteristics and explain how DiskSuite should be configured and used to take advantage of the SSA-100. This appendix assumes that you have Version 2.0 of the SPARCstorage Array software.

For situations not covered in this appendix, please refer to the SPARCstorage Array User's Guide for more information.

Installation

The SSA-100 should be installed according to the instructions in the SSA User's Guide. In particular, when using Solaris 2.4 or earlier releases, the SSA drivers contained in the SSA software distribution must be installed prior to using the SSA-100 with DiskSuite. The SSA Volume Manager need not be installed if you are only using DiskSuite.

SSA Specific Operations

Solstice DiskSuite 4.0 does not currently feature a graphical interface for accessing the SSA specific operations. To use these operations, you must use the SSA command line interface (CLI) provided with the SSA software distribution. Please refer to the sections describing SSA specific CLI operations in the SSA User's Guide for instructions on how to use these commands.

Device Naming

DiskSuite accesses SSA-100 disks exactly like any other disks, with one important exception: the disk names differ from non-SSA disks.

The SSA-100 disk naming convention is:

c[0-n]t[0-5]d[0-4]s[0-7]

where:

c indicates the controller attached to an SSA unit
t indicates one of the 6 SCSI strings within an SSA
d indicates one of the 5 disks on an internal SCSI string
s indicates the disk slice number
Strings t0 and t1 are contained in tray 1, t2 and t3 in tray 2, and t4 and t5 are in tray 3

Configuring for Online Replacement

If you want to take advantage of the SSA-100's online replacement capabilities, you must use either mirrors or RAID metadevices in order to provide data redundancy. The following sections describe how to set up mirrors and RAID devices so that failed drives can be replaced without loss of service.

Mirrors

When setting up mirrors on an SSA-100, you should configure the submirrors within a mirror from disks in different trays. This way, if a disk within a submirror fails, it will be possible to remove the tray and replace the drive without impacting the other submirror.

For example, consider a two-way mirror with each submirror composed of a concatenation of three SSA-100 disks. The following commands initialize this mirror using the first three drives from tray 1 and the first three drives in tray 2.


  # metainit d1 3 1 c0t0d0s2 1 c0t0d1s2 1 c0t0d2s2  
  # metainit d2 3 1 c0t2d0s2 1 c0t2d1s2 1 c0t2d2s2  
  # metainit d0 -m d1  
  # metattach d0 d2

Replacing a Disk Drive in a Mirror

Use the following steps to replace a SPARCstorage Array 100 drive when using DiskSuite. Where appropriate, the instructions give the DiskSuite CLI command required to perform the task. Each of these steps may also be performed using DiskSuite Tool. In this case, use the DiskSuite Tool operation corresponding to the given CLI command. You may want to consult the DiskSuite Tool User's Guide if you are unsure of which graphical operation corresponds to a given command.

Note - The following procedure is for Solaris 2.4 and previous Solaris versions only.

Use the metaoffline DiskSuite command to instruct DiskSuite to stop using all drives in the tray containing the failed drive. This may require offlining drives in submirrors unrelated to the submirror being repaired.

Warning - Any applications using non-replicated disks in this tray should now be suspended or terminated.

Use the ssacli sync_cache <ctlr> SSA command to flush outstanding SSA writes in NVRAM.
Use the ssacli -t<1|2|3> stop <ctlr> SSA command to spin down all drives in the tray containing the failed drive.
Remove the tray containing the failed drive and replace the drive.

Use the ssacli -t<1|2|3> start <ctlr> SSA command to spin up the drives in the tray.
Using format(1M) or fmthard(1M), correctly partition the new drive.
Use the metaonline DiskSuite command to instruct DiskSuite to start using the drives in the tray again.
Use the metareplace -e command to enable the new drive.

RAID-5 Metadevices

When setting up RAID-5 metadevices for online repair, you will have to use a minimum RAID-5 width of 3 components. While this is not an optimal configuration for RAID-5, it is still slightly less expensive than mirroring, in terms of the overhead of the redundant data. You should place each of the 3 components of each RAID-5 metadevice within a separate tray. If all disks in an SSA-100 are configured this way (or in combination with mirrors as described above), the tray containing the failed component may be removed without losing access to any of the data.

Replacing a Disk Drive in a RAID-5 Metadevice

Warning - Any applications using non-replicated disks in the tray containing the failed drive should first be suspended or terminated.

Use the ssacli sync_cache <ctlr> SSA command to flush outstanding SSA writes in NVRAM.
Use the ssacli -t<1|2|3> stop <ctlr> SSA command to spin down all drives in the tray containing the failed drive.
Remove the tray containing the failed drive and replace the drive.
Use the ssacli -t<1|2|3> start <ctlr> SSA command to spin up the drives in the tray.
Using format(1M) or fmthard(1M), correctly partition the new drive.
Use the metareplace -e command to enable the new drive in the tray.