ssd

- Fibre Channel Arbitrated Loop disk device driver

Synopsis

ssd@port,target:partition

Description

The ssd driver supports Fibre Channel disk devices.

The specific type of each disk is determined by the SCSI inquiry command and reading the volume label stored on block 0 of the drive. The volume label describes the disk geometry and partitioning; it must be present or the disk cannot be mounted by the system.

The block-files access the disk using the system's normal buffering mechanism and are read and written without regard to physical disk records. A “raw” interface provides for direct transmission between the disk and the read or write buffer. A single read or write call usually results in one I/O operation; raw I/O is therefore more efficient when many bytes are transmitted. Block file names are found in /dev/dsk; the names of the raw files are found in /dev/rdsk.

I/O requests (such as lseek(2)) to the SCSI disk must have an offset that is a multiple of 512 bytes (DEV_BSIZE), or the driver returns an EINVAL error. If the transfer length is not a multiple of 512 bytes, the transfer count is rounded up by the driver.

Partition 0 is normally used for the root file system on a disk, with partition 1 as a paging area (for example, swap). Partition 2 is used to back up the entire disk. Partition 2 normally maps the entire disk and may also be used as the mount point for secondary disks in the system. The rest of the disk is normally partition 6. For the primary disk, the user file system is located here.

The device has associated error statistics. These must include counters for hard errors, soft errors and transport errors. Other data may be implemented as required.

DEVICE STATISTICS SUPPORT

The device maintains I/O statistics for the device and for partitions allocated for that device. For each device/partition, the driver accumulates reads, writes, bytes read, and bytes written. The driver also initiates hi-resolution time stamps at queue entry and exit points to enable monitoring of residence time and cumulative residence-length product for each queue.

Not all device drivers make per-partition IO statistics available for reporting. ssd and sd(7D) per-partition statistics are enabled by default but may be disabled in their configuration files.

ioctls

Refer to dkio(7I).

Errors

EACCES

Permission denied.

EBUSY

The partition was opened exclusively by another thread.

EFAULT

The argument was a bad address.

EINVAL

Invalid argument.

EIO

An I/O error occurred.

ENOTTY

The device does not support the requested ioctl function.

ENXIO

When returned during open(2), this error indicates the device does not exist.

EROFS

The device is a read-only device.

CONFIGURATION

You configure the ssd driver by defining properties in the ssd.conf file. The ssd driver supports the following properties:

enable-partition-kstats

The default value is 1, which causes partition IO statistics to be maintained. Set this value to zero to prevent the driver from recording partition statistics. This slightly reduces the CPU overhead for IO, mimimizes the amount of sar(1) data collected and makes these statistics unavailable for reporting by iostat(1M) even though the -p/-P option is specified. Regardless of this setting, disk IO statistics are always maintained.

In addition to the above properties, some device-specific tunables can be configured in ssd.conf using the 'ssd-config-list' global property. The value of this property is a list of duplets. The formal syntax is:

ssd-config-list =     <duplet> [, <duplet> ]* ;

where 

<duplet>:=        "<vid+pid>" , "<tunable-list>"

and 

<tunable-list>:=   <tunable>  [, <tunable> ]*;
<tunable> =        <name> : <value>

The <vid+pid> is the string that is returned by the target device
on a SCSI inquiry command.

The <tunable-list> contains one or more tunables to apply to
all target devices with the specified <vid+pid>.

Each <tunable> is a <name> : <value> pair. Supported
tunable names are:
  
   delay-busy:    when busy, nsecs of delay before retry.
   
   retries-timeout:    retries to perform on an IO timeout.

physical-block-size

SCSI Disk drivers take this value as the physical block size of the disks which do not report valid physical block size. The value must be a power of two. If not specified, DEV_BSIZE(512 bytes) is implied.

Examples

The following is an example of a global ssd-config-list property:

  
   ssd-config-list =
      "SUN     T4", "delay-busy:600, retries-timeout:6",
      "SUN     StorEdge_3510", "retries-timeout:3";

Files

ssd.conf

Driver configuration file

/dev/dsk/cntndnsn

block files

/dev/rdsk/cntndnsn

raw files

cn

is the controller number on the system.

tn

7-bit disk loop identifier, such as switch setting

dn

SCSI lun n

sn

partition n (0-7)

See Also

sar(1), format(1M), iostat(1M), ioctl(2), lseek(2), open(2), read(2), write(2), scsi(4), driver.conf(4), cdio(7I), dkio(7I)

ANSI Small Computer System Interface-2 (SCSI-2)

ANSI X3.272-1996, Fibre Channel Arbitrated Loop (FC-AL)

Fibre Channel - Private Loop SCSI Direct Attach (FC-PLDA)

Diagnostics

Error for command '<command name>' Error Level: Fatal Requested Block <n>, 
Error  Block: <m>, Sense Key: <sense key name>, Vendor '<vendor name>': 
ASC = 0x<a> (<ASC name>), ASCQ = 0x<b>, FRU = 0x<c>

The command indicated by <command name> failed. The Requested Block is the block where the transfer started and the Error Block is the block that caused the error. Sense Key, ASC, and ASCQ information is returned by the target in response to a request sense command.

Check Condition on REQUEST SENSE

A REQUEST SENSE command completed with a check condition. The original command will be retried a number of times.

Label says <m> blocks Drive says <n> blocks

There is a discrepancy between the label and what the drive returned on the READ CAPACITY command.

Not enough sense information

The request sense data was less than expected.

Request Sense couldn't get sense data

The REQUEST SENSE command did not transfer any data.

Reservation Conflict

The drive was reserved by another initiator.

SCSI transport failed: reason 'xxxx' : {retrying|giving up} 

The host adapter has failed to transport a command to the target for the reason stated. The driver will either retry the command or, ultimately, give up.

Unhandled Sense Key <n>

The REQUEST SENSE data included an invalid sense key.

Unit not Ready. Additional sense code 0x<n>

The drive is not ready.

corrupt label - bad geometry

The disk label is corrupted.

corrupt label - label checksum failed

The disk label is corrupted.

corrupt label - wrong magic number

The disk label is corrupted.

device busy too long

The drive returned busy during a number of retries.

disk not responding to selection

The drive was probably powered down or died.

i/o to invalid geometry

The geometry of the drive could not be established.

incomplete read/write - retrying/giving up

There was a residue after the command completed normally.

logical unit not ready

The drive is not ready.

no bp for disk label

A bp with consistent memory could not be allocated.

no mem for property

Free memory pool exhausted.

no memory for disk label

Free memory pool exhausted.

no resources for dumping

A packet could not be allocated during dumping.

offline

Drive went offline; probably powered down.

requeue of command fails<n>

Driver attempted to retry a command and experienced a transport error.

ssdrestart transport failed <n>

Driver attempted to retry a command and experienced a transport error.

transfer length not modulo <n>

Illegal request size.

transport rejected <n>

Host adapter driver was unable to accept a command.

unable to read label

Failure to read disk label.

unit does not respond to selection

Drive went offline; probably powered down.