cciss.txt 6.5 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165
  1. This driver is for Compaq's SMART Array Controllers.
  2. Supported Cards:
  3. ----------------
  4. This driver is known to work with the following cards:
  5. * SA 5300
  6. * SA 5i
  7. * SA 532
  8. * SA 5312
  9. * SA 641
  10. * SA 642
  11. * SA 6400
  12. * SA 6400 U320 Expansion Module
  13. * SA 6i
  14. * SA P600
  15. * SA P800
  16. * SA E400
  17. * SA P400i
  18. * SA E200
  19. * SA E200i
  20. * SA E500
  21. If nodes are not already created in the /dev/cciss directory, run as root:
  22. # cd /dev
  23. # ./MAKEDEV cciss
  24. Device Naming:
  25. --------------
  26. You need some entries in /dev for the cciss device. The MAKEDEV script
  27. can make device nodes for you automatically. Currently the device setup
  28. is as follows:
  29. Major numbers:
  30. 104 cciss0
  31. 105 cciss1
  32. 106 cciss2
  33. 105 cciss3
  34. 108 cciss4
  35. 109 cciss5
  36. 110 cciss6
  37. 111 cciss7
  38. Minor numbers:
  39. b7 b6 b5 b4 b3 b2 b1 b0
  40. |----+----| |----+----|
  41. | |
  42. | +-------- Partition ID (0=wholedev, 1-15 partition)
  43. |
  44. +-------------------- Logical Volume number
  45. The device naming scheme is:
  46. /dev/cciss/c0d0 Controller 0, disk 0, whole device
  47. /dev/cciss/c0d0p1 Controller 0, disk 0, partition 1
  48. /dev/cciss/c0d0p2 Controller 0, disk 0, partition 2
  49. /dev/cciss/c0d0p3 Controller 0, disk 0, partition 3
  50. /dev/cciss/c1d1 Controller 1, disk 1, whole device
  51. /dev/cciss/c1d1p1 Controller 1, disk 1, partition 1
  52. /dev/cciss/c1d1p2 Controller 1, disk 1, partition 2
  53. /dev/cciss/c1d1p3 Controller 1, disk 1, partition 3
  54. SCSI tape drive and medium changer support
  55. ------------------------------------------
  56. SCSI sequential access devices and medium changer devices are supported and
  57. appropriate device nodes are automatically created. (e.g.
  58. /dev/st0, /dev/st1, etc. See the "st" man page for more details.)
  59. You must enable "SCSI tape drive support for Smart Array 5xxx" and
  60. "SCSI support" in your kernel configuration to be able to use SCSI
  61. tape drives with your Smart Array 5xxx controller.
  62. Additionally, note that the driver will not engage the SCSI core at init
  63. time. The driver must be directed to dynamically engage the SCSI core via
  64. the /proc filesystem entry which the "block" side of the driver creates as
  65. /proc/driver/cciss/cciss* at runtime. This is because at driver init time,
  66. the SCSI core may not yet be initialized (because the driver is a block
  67. driver) and attempting to register it with the SCSI core in such a case
  68. would cause a hang. This is best done via an initialization script
  69. (typically in /etc/init.d, but could vary depending on distribution).
  70. For example:
  71. for x in /proc/driver/cciss/cciss[0-9]*
  72. do
  73. echo "engage scsi" > $x
  74. done
  75. Once the SCSI core is engaged by the driver, it cannot be disengaged
  76. (except by unloading the driver, if it happens to be linked as a module.)
  77. Note also that if no sequential access devices or medium changers are
  78. detected, the SCSI core will not be engaged by the action of the above
  79. script.
  80. Hot plug support for SCSI tape drives
  81. -------------------------------------
  82. Hot plugging of SCSI tape drives is supported, with some caveats.
  83. The cciss driver must be informed that changes to the SCSI bus
  84. have been made, in addition to and prior to informing the SCSI
  85. mid layer. This may be done via the /proc filesystem. For example:
  86. echo "rescan" > /proc/scsi/cciss0/1
  87. This causes the adapter to query the adapter about changes to the
  88. physical SCSI buses and/or fibre channel arbitrated loop and the
  89. driver to make note of any new or removed sequential access devices
  90. or medium changers. The driver will output messages indicating what
  91. devices have been added or removed and the controller, bus, target and
  92. lun used to address the device. Once this is done, the SCSI mid layer
  93. can be informed of changes to the virtual SCSI bus which the driver
  94. presents to it in the usual way. For example:
  95. echo scsi add-single-device 3 2 1 0 > /proc/scsi/scsi
  96. to add a device on controller 3, bus 2, target 1, lun 0. Note that
  97. the driver makes an effort to preserve the devices positions
  98. in the virtual SCSI bus, so if you are only moving tape drives
  99. around on the same adapter and not adding or removing tape drives
  100. from the adapter, informing the SCSI mid layer may not be necessary.
  101. Note that the naming convention of the /proc filesystem entries
  102. contains a number in addition to the driver name. (E.g. "cciss0"
  103. instead of just "cciss" which you might expect.)
  104. Note: ONLY sequential access devices and medium changers are presented
  105. as SCSI devices to the SCSI mid layer by the cciss driver. Specifically,
  106. physical SCSI disk drives are NOT presented to the SCSI mid layer. The
  107. physical SCSI disk drives are controlled directly by the array controller
  108. hardware and it is important to prevent the kernel from attempting to directly
  109. access these devices too, as if the array controller were merely a SCSI
  110. controller in the same way that we are allowing it to access SCSI tape drives.
  111. SCSI error handling for tape drives and medium changers
  112. -------------------------------------------------------
  113. The linux SCSI mid layer provides an error handling protocol which
  114. kicks into gear whenever a SCSI command fails to complete within a
  115. certain amount of time (which can vary depending on the command).
  116. The cciss driver participates in this protocol to some extent. The
  117. normal protocol is a four step process. First the device is told
  118. to abort the command. If that doesn't work, the device is reset.
  119. If that doesn't work, the SCSI bus is reset. If that doesn't work
  120. the host bus adapter is reset. Because the cciss driver is a block
  121. driver as well as a SCSI driver and only the tape drives and medium
  122. changers are presented to the SCSI mid layer, and unlike more
  123. straightforward SCSI drivers, disk i/o continues through the block
  124. side during the SCSI error recovery process, the cciss driver only
  125. implements the first two of these actions, aborting the command, and
  126. resetting the device. Additionally, most tape drives will not oblige
  127. in aborting commands, and sometimes it appears they will not even
  128. obey a reset command, though in most circumstances they will. In
  129. the case that the command cannot be aborted and the device cannot be
  130. reset, the device will be set offline.
  131. In the event the error handling code is triggered and a tape drive is
  132. successfully reset or the tardy command is successfully aborted, the
  133. tape drive may still not allow i/o to continue until some command
  134. is issued which positions the tape to a known position. Typically you
  135. must rewind the tape (by issuing "mt -f /dev/st0 rewind" for example)
  136. before i/o can proceed again to a tape drive which was reset.