Documentation/cciss.txt

   1 This driver is for Compaq's SMART Array Controllers.
   2
   3 Supported Cards:
   4 ----------------
   5
   6 This driver is known to work with the following cards:
   7
   8         * SA 5300
   9         * SA 5i
  10         * SA 532
  11         * SA 5312
  12         * SA 641
  13         * SA 642
  14         * SA 6400
  15         * SA 6400 U320 Expansion Module
  16         * SA 6i
  17         * SA P600
  18         * SA P800
  19         * SA E400
  20         * SA P400i
  21         * SA E200
  22         * SA E200i
  23         * SA E500
  24
  25 Detecting drive failures:
  26 -------------------------
  27
  28 To get the status of logical volumes and to detect physical drive
  29 failures, you can use the cciss_vol_status program found here:
  30 http://cciss.sourceforge.net/#cciss_utils
  31
  32 Device Naming:
  33 --------------
  34
  35 If nodes are not already created in the /dev/cciss directory, run as root:
  36
  37 # cd /dev
  38 # ./MAKEDEV cciss
  39
  40 You need some entries in /dev for the cciss device.  The MAKEDEV script
  41 can make device nodes for you automatically.  Currently the device setup
  42 is as follows:
  43
  44 Major numbers:
  45         104     cciss0
  46         105     cciss1
  47         106     cciss2
  48         105     cciss3
  49         108     cciss4
  50         109     cciss5
  51         110     cciss6
  52         111     cciss7
  53
  54 Minor numbers:
  55         b7 b6 b5 b4 b3 b2 b1 b0
  56         |----+----| |----+----|
  57              |           |
  58              |           +-------- Partition ID (0=wholedev, 1-15 partition)
  59              |
  60              +-------------------- Logical Volume number
  61
  62 The device naming scheme is:
  63 /dev/cciss/c0d0                 Controller 0, disk 0, whole device
  64 /dev/cciss/c0d0p1               Controller 0, disk 0, partition 1
  65 /dev/cciss/c0d0p2               Controller 0, disk 0, partition 2
  66 /dev/cciss/c0d0p3               Controller 0, disk 0, partition 3
  67
  68 /dev/cciss/c1d1                 Controller 1, disk 1, whole device
  69 /dev/cciss/c1d1p1               Controller 1, disk 1, partition 1
  70 /dev/cciss/c1d1p2               Controller 1, disk 1, partition 2
  71 /dev/cciss/c1d1p3               Controller 1, disk 1, partition 3
  72
  73 SCSI tape drive and medium changer support
  74 ------------------------------------------
  75
  76 SCSI sequential access devices and medium changer devices are supported and
  77 appropriate device nodes are automatically created.  (e.g.
  78 /dev/st0, /dev/st1, etc.  See the "st" man page for more details.)
  79 You must enable "SCSI tape drive support for Smart Array 5xxx" and
  80 "SCSI support" in your kernel configuration to be able to use SCSI
  81 tape drives with your Smart Array 5xxx controller.
  82
  83 Additionally, note that the driver will not engage the SCSI core at init
  84 time.  The driver must be directed to dynamically engage the SCSI core via
  85 the /proc filesystem entry which the "block" side of the driver creates as
  86 /proc/driver/cciss/cciss* at runtime.  This is because at driver init time,
  87 the SCSI core may not yet be initialized (because the driver is a block
  88 driver) and attempting to register it with the SCSI core in such a case
  89 would cause a hang.  This is best done via an initialization script
  90 (typically in /etc/init.d, but could vary depending on distribution).
  91 For example:
  92
  93         for x in /proc/driver/cciss/cciss[0-9]*
  94         do
  95                 echo "engage scsi" > $x
  96         done
  97
  98 Once the SCSI core is engaged by the driver, it cannot be disengaged
  99 (except by unloading the driver, if it happens to be linked as a module.)
 100
 101 Note also that if no sequential access devices or medium changers are
 102 detected, the SCSI core will not be engaged by the action of the above
 103 script.
 104
 105 Hot plug support for SCSI tape drives
 106 -------------------------------------
 107
 108 Hot plugging of SCSI tape drives is supported, with some caveats.
 109 The cciss driver must be informed that changes to the SCSI bus
 110 have been made, in addition to and prior to informing the SCSI
 111 mid layer.  This may be done via the /proc filesystem.  For example:
 112
 113         echo "rescan" > /proc/scsi/cciss0/1
 114
 115 This causes the adapter to query the adapter about changes to the
 116 physical SCSI buses and/or fibre channel arbitrated loop and the
 117 driver to make note of any new or removed sequential access devices
 118 or medium changers.  The driver will output messages indicating what
 119 devices have been added or removed and the controller, bus, target and
 120 lun used to address the device.  Once this is done, the SCSI mid layer
 121 can be informed of changes to the virtual SCSI bus which the driver
 122 presents to it in the usual way. For example:
 123
 124         echo scsi add-single-device 3 2 1 0 > /proc/scsi/scsi
 125
 126 to add a device on controller 3, bus 2, target 1, lun 0.   Note that
 127 the driver makes an effort to preserve the devices positions
 128 in the virtual SCSI bus, so if you are only moving tape drives
 129 around on the same adapter and not adding or removing tape drives
 130 from the adapter, informing the SCSI mid layer may not be necessary.
 131
 132 Note that the naming convention of the /proc filesystem entries
 133 contains a number in addition to the driver name.  (E.g. "cciss0"
 134 instead of just "cciss" which you might expect.)
 135
 136 Note: ONLY sequential access devices and medium changers are presented
 137 as SCSI devices to the SCSI mid layer by the cciss driver.  Specifically,
 138 physical SCSI disk drives are NOT presented to the SCSI mid layer.  The
 139 physical SCSI disk drives are controlled directly by the array controller
 140 hardware and it is important to prevent the kernel from attempting to directly
 141 access these devices too, as if the array controller were merely a SCSI
 142 controller in the same way that we are allowing it to access SCSI tape drives.
 143
 144 SCSI error handling for tape drives and medium changers
 145 -------------------------------------------------------
 146
 147 The linux SCSI mid layer provides an error handling protocol which
 148 kicks into gear whenever a SCSI command fails to complete within a
 149 certain amount of time (which can vary depending on the command).
 150 The cciss driver participates in this protocol to some extent.  The
 151 normal protocol is a four step process.  First the device is told
 152 to abort the command.  If that doesn't work, the device is reset.
 153 If that doesn't work, the SCSI bus is reset.  If that doesn't work
 154 the host bus adapter is reset.  Because the cciss driver is a block
 155 driver as well as a SCSI driver and only the tape drives and medium
 156 changers are presented to the SCSI mid layer, and unlike more
 157 straightforward SCSI drivers, disk i/o continues through the block
 158 side during the SCSI error recovery process, the cciss driver only
 159 implements the first two of these actions, aborting the command, and
 160 resetting the device.  Additionally, most tape drives will not oblige
 161 in aborting commands, and sometimes it appears they will not even
 162 obey a reset command, though in most circumstances they will.  In
 163 the case that the command cannot be aborted and the device cannot be
 164 reset, the device will be set offline.
 165
 166 In the event the error handling code is triggered and a tape drive is
 167 successfully reset or the tardy command is successfully aborted, the
 168 tape drive may still not allow i/o to continue until some command
 169 is issued which positions the tape to a known position.  Typically you
 170 must rewind the tape (by issuing "mt -f /dev/st0 rewind" for example)
 171 before i/o can proceed again to a tape drive which was reset.
 172