1. DL manuals
  2. IBM
  3. Server
  4. Multiprise 3000
  5. Reference Manual

IBM Multiprise 3000 Reference Manual - Contents

Manual is about: Enterprise Server Internal Disk Subsystem

Summary of Multiprise 3000

  • Page 1

    S/390 multiprise 3000 enterprise server r ibm internal disk subsystem: reference guide sa22-1025-00

  • Page 3

    S/390 ibm internal disk subsystem: reference guide sa22-1025-00

  • Page 4

    Note! Before using this information and the product it supports, be sure to read the general information under appendix d, “notices” on page d-1. First edition (november, 1999) this edition, sa22-1025-00, applies to the ibm internal disk subsystem.  copyright international business machines corpora...

  • Page 5: Contents

    Contents chapter 1. Overview of the s/390 internal disk subsystem . . . . . . . . . 1-1 host processor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-2 raid disk arrays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1-3 hdd characteristics . . . ...

  • Page 6

    End-of-file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-4 chapter 4. Command descriptions . . . . . . . . . . . . . . . . . . . . . . . . 4-1 channel command word . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-1 exception conditions . . . . . . . . ...

  • Page 7

    Write track data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-102 sense commands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-104 sense . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-104 sense id . . . . . . . . ....

  • Page 8

    Sense byte 5–device type code . . . . . . . . . . . . . . . . . . . . . . . . . 6-6 sense byte 6–content and format . . . . . . . . . . . . . . . . . . . . . . . . 6-6 sense bytes 7 through 19 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6-6 sense bytes 20 and 21–subsystem identifier . ...

  • Page 9

    Chapter 1. Overview of the s/390 internal disk subsystem this chapter gives an overview of the s/390 internal disk subsystem, describes its components, features, and facilities. Internal disk storage subsystem provides internal attachment of disk storage integrated into the central processing comple...

  • Page 10

    96 for an array consisting of 7 hdds (6+p). Also, logical volumes can be added incrementally. This can even be done while i/o operations are ongoing to other logical volumes which already exist on this array. In the same way, logical volumes can also be selectively deleted again, freeing up space wh...

  • Page 11

    Power supplies. Two cooling fans are also provided to assure that the power and cooling system is fully redundant with concurrent repair. An integrated battery feature (ibf) is available. Internal disk in the multiprise 3000 continues the evolution of disk capacity with the introduction of raid disk...

  • Page 12

    Drive is mounted on a carrier which facilitates installation into the hdd enclosure and provides additional shock protection for the hdd. The carrier includes 3 leds: – green led indicating good power – another green led indicating activity – amber led indicating a check other major disk drive param...

  • Page 13

    Ckd emulation the internal disk subsystem processes eckd/ckd channel programs which presuppose variable length count, key, and data fields like those of a 3990 subsystem. Internal disk constructs ckd track images in memory and executes the eckd and ckd channel programs, which operate on byte boundar...

  • Page 14

    Background scrubbing background scrubbing is provided by internal disk to further reduce the probability of customer loss of data. There is a small possibility that after a data failure occurs, the redundancy data is also found to be unreadable, resulting in loss of data to the customer. Data scrubb...

  • Page 15

    Executed when a loop is detected. With only one transfer of command parameters from the application program, either the desired record is found or an exception condition occurs. A second transfer of parameters may then be needed to structure the ending conditions for the loop, such as providing read...

  • Page 16

    Total storage, not on 1gb and 2gb models. The size can be changed by the operator in 32mb increments. Ssa adapter card the ssa adapter card supplies an ssa interface that operates at up to 40mb/sec for connection of the hdds. Two levels of physical interface conversion occur between the multiprise 3...

  • Page 17

    Logical volumes are built. The number of storage units per array depends on the size of the array. For the supported array sizes, these are the corresponding number of storage units: the following types of logical volumes can be mapped to an array, each requiring the number of storage units shown: t...

  • Page 18

    Control unit logical address each of the three enclosures of a fully populated multiprise 3000 contains a disk subsystem with a logical 3990 mod 2 control unit. The logical control unit that controls the hdd enclosure in the cpc frame is assigned control unit logical address x'0'. The logical contro...

  • Page 19

    Table 1-5. Configurations in cpc frame number configuration approximate gross disk capacity (gbyte) e1 4+p, spare 4 * 18 = 72 e2 4+p, 6+p, spare 10 * 18 = 180 e3 4+p, 6+p, 6+p, spare 16 * 18 = 288 commands and facilities not supported internal disk does not support the following commands and facilit...

  • Page 20

    Defined for each raid-5 array. Each logical volume may only be defined to one control unit. The self timed interface (sti) connection between the processor and the ssa adapter is specified as the single dsd chpid. Control unit (cntlunit) statements are defined for each ssa loop in the storage facili...

  • Page 21

    Chapter 2. Processing commands and presenting status an understanding of the enterprise systems architecture/390  channels is assumed throughout this manual. The only points discussed are those points requiring further clarity, or where the subsystem has an option. The following i/o instructions in...

  • Page 22

    Table 2-1. Status byte bit name 0 attention 1 status modifier 2 control-unit end 3 busy 4 channel end 5 device end 6 unit check 7 unit exception attention (bit 0) bit 0 is set to a ‘1’: with device end and unit exception to show a state transition. (see “state-change interruption” on page 2-5.) with...

  • Page 23

    Channel end (bit 4) bit 4 is set to a ‘1’: at the end of parameter transfer or data transfer of each command. At the end of command execution for commands that have no parameter or data transfer. Device end (bit 5) bit 5 is set to a ‘1’: when alone or with channel end to show that a logical volume i...

  • Page 24

    Initial status byte the initial status byte is zero for all valid non-immediate commands with the following exceptions: logical volume is busy – busy bit alone is returned. Status condition is pending – the pending status combination plus busy bit is returned. See “status pending conditions.” busy s...

  • Page 25

    Channel end has been previously accepted and the operation is now complete, but device end has not been presented to the channel. Busy status has previously been sent and the logical volume is now not busy. The resulting device end has not been presented to the channel. The logical volume changed st...

  • Page 26

    Chain. These command chains may compete for data within a single logical volume. Only one data transfer chain is active on a logical volume at a time. I/o queuing multiple lpars may issue ccw chains that require the same logical volume. Internal disk provides for queuing of command chains from the v...

  • Page 27

    Table column definitions the table that follows is intended to show the actions taken for the bypass commands particularly. Table 2-2 on page 2-8 summarizes the status presented to commands when the subsystem or device is in one of the states described in the following text. Hdd long busy. This stat...

  • Page 28

    Command mnemonic nop spid snid ur rsta ripl res rel sns (1) sns (2) sns (3) snss rdc snsid rcd psf rssd smr all other commands subsystem contingent allegiance different lpar accepted accepted accepted accepted accepted accepted accepted accepted accepted accepted accepted accepted accepted accepted ...

  • Page 29

    Listing of commands appearing in table 3 a full listing of all of the command names and mnemonics that are executed by internal disk appears in table 7 on page 4-3. For convenience in referring to the preceding table a list of only the commands that appear in the table is shown below. The commands a...

  • Page 30

    Uses channel command retry to disconnect from the channel and recover from the error or wait for the required resource. Channel command retry–control unit initiated the control unit initiates retries for delayed selection and device orientation. Delayed selection and device orientation internal disk...

  • Page 31

    Alone subsystem that has external interfaces and interface switches in its attachment topology. The first selection of all other logical volumes on the reset interface will be unit checked with sense data that includes equipment check. The internal disk subsystem procedure is that the first selectio...

  • Page 32

    2-12 internal disk subsystem reference guide.

  • Page 33

    Chapter 3. Elements of data access the internal disk subsystem emulates the characteristics of 3380 and 3390 devices while using real raid arrays for disk storage. References to devices in this section and all sections in this publication which are concerned with a view of data, refer to the perspec...

  • Page 34

    Track accessing to access a specific track, the host program must first specify the desired cylinder and track. This positioning occurs without requiring the device to read any user data. When the track accessing operation completes, the track address (cchh) is saved and maintained by the subsystem ...

  • Page 35

    Record accessing an associative addressing (search) operation can find a specific record, which causes the control unit to read some area from the device and compare (associate) it with a value from the channel. The locate record command or one of the search commands is used to do the search operati...

  • Page 36

    See “locate record” on page 4-13 for a complete description. Note: performance can be significantly affected by the way that locate record ccws are used in the channel program. For example, two different channel programs can read two consecutive records on a track. One channel program has one locate...

  • Page 37

    Page a-1. See also ibm 3380 direct access storage introduction and ibm 3390 direct access storage introduction. Chapter 3. Elements of data access 3-5.

  • Page 38

    3-6 internal disk subsystem reference guide.

  • Page 39

    Chapter 4. Command descriptions this section describes the channel command word and the valid channel commands for the internal disk. Channel command word the channel command word (ccw) specifies the command, and designates the processor storage area associated with the operation and the action take...

  • Page 40

    Indirect data address (ida): bit 37 (format 0) or bit 13 (format 1), when ‘1’, specifies indirect data addressing. Suspend (s): bit 38 (format 0) or bit 14 (format 1), when ‘1’, specifies suspension of channel-program processing. Note: except for a transfer in channel (tic) command, flag bit 39 (for...

  • Page 41

    If the next track is not in the defined extent, the operation ends with unit check status. The sense data includes file protected. If the operation is a locate record extended operation code of read any or write any, the operation continues through index on the same track. Operations outside the dom...

  • Page 42

    Table 4-3 (page 2 of 3). List of channel commands command name mnemonic hex code page single-track mode multitrack mode read commands read home address read record zero read count, key, and data read key and data read data read count read multiple count, key, and data read track read track data read...

  • Page 43

    Table 4-3 (page 3 of 3). List of channel commands command name mnemonic hex code page single-track mode multitrack mode subsystem commands perform subsystem function sense subsystem status read subsystem data psf snss rssd 27 54 3e – – – 4-130 4-136 4-139 note: (1) this command is valid on 3380 logi...

  • Page 44

    List of channel commands by hex code following is a list of the same commands as in table 4-3 on page 4-3. The commands are listed here in order by their hex codes, with the page location of the command descriptions. (mt) = multitrack mode. Hex command name page 02 read ipl 4-76 03 no-operation 4-10...

  • Page 45

    Addressing and control commands none of the commands in this group transfer user data records between the internal disk and the host channel. However, most of the commands in this group require the host system to send control information, associated with the command, to the internal disk. Control in...

  • Page 46

    Define extent description: the define extent command parameters define limits on operations that follow, provide a blocksize value, and specify cache controls for the channel program. This command resets orientation in the control unit. Chaining and restrictions: unit check status with the sense dat...

  • Page 47

    Byte 0, bits 0 and 1: specify what types of write operations to permit in the channel program. The 3390 has a restriction: note: "if these bits are ‘11’, byte 1, bits 3 through 5 are ignored and bypass cache is forced." this restriction assures that write ha and write r0 commands that can alter the ...

  • Page 48

    Byte 0, bits 3 and 4: specify what seek commands and multitrack head switching operations to permit outside the domain of a locate record or locate record extended. They have no effect on channel commands operating in the domain of a locate record or locate record extended. Byte 0, bits 5 and 6: spe...

  • Page 49

    Byte 1, bits 0 and 1 – eckd mode: when ‘11’, specify eckd mode. The internal disk supports only the eckd mode of the define extent command. If bits 0 and 1 are not ‘11’, the command is rejected with unit check status. The sense data contains command reject with format 0, message 4. Byte 1, bit 2 – c...

  • Page 50

    ‘100’ – sequential prestage: in the internal disk subsystem, this setting is treated the same as normal cache replacement. ‘101’ – record access: indicates that there is little locality of reference of the data accessed by this channel program. Byte 1, bit 6 – reserved, must be zero: if this bit is ...

  • Page 51

    The data field bytes are all equal to zero. If bit 5 is set to zero, the record zero differs from the above only in the data field content. The data field bytes are those that are provided in the parameters of the write record zero command. Therefore, the data field bytes may or may not be equal to ...

  • Page 52

    The control unit verifies the parameters and reserved fields in their order in table 4-7 on page 4-14, except it checks byte 0, bits 2 through 7 first. If the control unit detects an exception condition in more than one parameter in table 4-7 on page 4-14, it reports only the first exception. Other ...

  • Page 53

    Orientation modifiers: byte 0, bits 0 and 1 contain orientation modifiers that specify the orientation to establish when the operation (bits 2 through 7) needs orientation to a specific record or record area before data transfer starts. Count area ‘00’ – specifies that data access for the ccw immedi...

  • Page 54

    Valid operation byte values:: the following table shows the valid combinations of orientation modifiers and operation codes. All of the valid combinations are listed. Table 4-9. Locate record operation byte values operation code (byte 0, bits 2–7) orientation (byte 0, bits 0–1) cnt ha data idx orien...

  • Page 55

    Other code, the command is rejected with unit check status. The sense data contains command reject with format 0, message 4. Not used – byte 2 byte 2 must contain zeros. If byte 2 does not contain zeros, the locate record command is rejected with unit check status. The sense data contains command re...

  • Page 56

    If the sector number is zero, the next command starts with the device in a position immediately before index. A value of x'ff' is valid and specifies that sector positioning is not to be performed before establishing orientation. Transfer length factor – bytes 14 and 15 when the auxiliary byte, byte...

  • Page 57

    When operating with read intent, an incomplete domain exception will not be reported. The internal disk subsystem will threshold unit checks for incomplete domain exceptions. See “nonsynchronous operations” on page 4-19. Search operation: certain locate record operations need orientation to a specif...

  • Page 58

    Data transfer to the disk subsystem. If the exception condition is related to the execution of a write command that had already presented clean ending status, sense data may also include imprecise ending (byte 1, bit 7). If an incomplete domain exception condition is detected after final status is p...

  • Page 59

    The control unit uses the seek address and sector number parameters to start track access. When track access completes, the search operation specified by the search argument and the orientation modifiers (byte 0, bits 0 and 1) is performed. If the auxiliary byte bit 7 is ‘0’, the locate record comma...

  • Page 60

    If the record length is not zero, the operation is terminated with unit check status. Sense data will include program action code ‘0f’. Format write – ‘03’ operation code: this operation code prepares the control unit to format one or more consecutive user data records. The count parameter specifies...

  • Page 61

    The control unit uses the seek address and sector number parameters to start track access. When track access is complete, the search operation specified by the search argument and the orientation modifiers (byte 0, bits 0 and 1) is performed. The locate record command must be followed by the number ...

  • Page 62

    The locate record command must be followed by one write data and a number of write ckd commands that is one less than the value in the count parameter, with the following exceptions: an erase command may be substituted for the last write ckd command in the domain. If the count parameter contains a v...

  • Page 63

    After the last record on a track transfers, a pseudo count field of 8 bytes of x'ff' transfers to the channel. If the track has no records to transfer, the read tracks command sends only the pseudo count field. Note: the pseudo count field can be used to locate the end of the track image character s...

  • Page 64

    Only read count, read key and data, read data, or read count key and data commands can follow a read r0, read count, read key and data, read data, or a read count key and data command. If a command in the locate record domain does not meet the sequence requirements, that command is rejected with uni...

  • Page 65

    Locate record extended description: the location and number of records to process and the operation to perform are specified in the locate record extended command parameters. See page 3-3 for how the locate record extended command makes an operational domain. See “locate record extended operations” ...

  • Page 66

    Other restrictions are described in the following text. Ending status: the control unit presents channel end status after it verifies the parameters. If the operation uses cache, and the track and sector specified in the locate record extended command is in cache, device end occurs with channel end....

  • Page 67

    Orientation modifiers: byte 0, bits 0 and 1 contain orientation modifiers that specify the orientation to establish when the operation (bits 2 through 7) needs orientation to a specific record or record area before data transfer starts. Count area ‘00’ – specifies that data access for the command im...

  • Page 68

    See "locate record extended operations" on page 4-26 for a description of the operation codes. Valid operation byte values: the following table shows the valid combinations of orientation modifiers and operation codes. All of the valid combinations are listed. Table 4-13. Locate record extended oper...

  • Page 69

    Byte 1, bit 7: when set to ‘1’, specifies that the last data transfer command in the locate record extended domain will be a read count command. The read count command is a suffix to the locate record extended domain. The record is included in the count of records processed as specified by the count...

  • Page 70

    Search argument – bytes 8–12 bytes 8 through 12 specify the cylinder, head, and record (cchhr). These bytes specify a value to be used as a search argument for the locate record extended search operation. See “search operation” on page 4-19. When the operation (byte 0) does not need orientation to a...

  • Page 71

    Extended operation – byte 17 byte 17 contains an operation code unique to the locate record extended command. When byte 0, bits 2–7 equal x'3f', byte 17 specifies the locate record extended operation to be performed. If byte 0, bits 2–7 is not x'3f' the operation specified in byte 0, bits 2–7 is per...

  • Page 72

    Table 4-16. Extended operation code length extended operation codes length write any (09) x'0001' read any (0a) x'0001' read trackset (0e) x'0001' or x'0002' prestage trackset (10) x'0001' or x'0002' write trackset (11) x'0001' or x'0002' update write trackset (13) x'0001' or x'0002' extended parame...

  • Page 73

    Command is rejected with unit check status. The sense data contains command reject with format 0, message 4. Notes: 1. Extent checking is performed during the validation of locate record extended parameters, not during track access. 2. The sector number parameter is not used. Track access is initiat...

  • Page 74

    The extended parameter field must be one byte in length (byte 20 only). This byte specifies the set size used in the locate record extended domain. The track set size must be ‘1’ (the device value returned in byte 47 of the read device characteristics command response). If the track set size byte is...

  • Page 75

    If the number of records written is equal to the number of data records on the track, the count area read by a suffixed read count command is that of the first record that was updated. If the number of records written is not equal to the number of data records on the track, the count area read by a ...

  • Page 76

    Each following read track command attempts to advance to the next specified track, orients to the home address, and transfers each count, key, and data area until end-of-track. A pseudo count field is transferred to the channel after the last record on a track is transferred. If no records were tran...

  • Page 77

    Write trackset – ‘11’ extended operation code: the write trackset operation prepares the control unit to update the record zero data area and format write the user data records on one or more tracks. The tracks to be written are specified by the extended parameter and the number of tracks to be writ...

  • Page 78

    Format 0, message 2, invalid command sequence). An invalid command sequence indication will be returned for any non-conforming command even if it is a command that is not executed by internal disk. For example, the read backward command, ‘0c’, that is executed on tape devices and is never executed b...

  • Page 79

    Disk, would encounter invalid command sequence rather than invalid command in this instance. Chapter 4. Command descriptions 4-41.

  • Page 80

    Prefix hex code data address count e7 the main storage location of the first prefix parameter byte. Minimum of 64. Description: the prefix command consolidates many commands that are used to convey information to the internal disk subsystem. This command reduces the channel overheads associated with...

  • Page 81

    Format byte: field validity byte: auxiliary byte: the prefix command has different formats that are described in the following sections. Table 4-17. Prefix, byte 0, format bits value definition 0–7 encoded format 00 basic prefix command (includes define extent) 01 basic plus locate record extended p...

  • Page 82

    Basic prefix command parameters - format byte x‘00’ the parameter list has the following format: byte parameter 0 format 1 flags 2 not used 3 auxiliary byte 4–11 not used 12–43 define extent parameters qualified by byte 1, bit 0. 12 mask byte 13 global attributes 14–15 blocksize in bytes 16–17 ignor...

  • Page 83

    Byte parameter 62–63 perform subsystem function common bytes 44 operation byte 62 psf order 63 psf flags 64–nn psf order unique parameters. Nn is dependent on the order and can be 65535. Chapter 4. Command descriptions 4-45.

  • Page 84

    Seek operations when the control unit receives a seek, seek cylinder, or seek head command, it validates the seek address parameter, determines if the track is in cache, and saves the parameters until it requires data access. Further activity is deferred as described in the following text. While a s...

  • Page 85

    Seek description: the seek command transfers the track address to the control unit. The format of the seek address parameter is (00cchh), where (00) is 2 bytes of binary zeros and (cchh) is a 4-byte track address. The control unit saves the seek address to do a cache directory search or to position ...

  • Page 86

    Seek cylinder description: the seek cylinder command transfers the track address to the control unit. The format of the seek address parameter is (00cchh), where (00) is 2 bytes of binary zeros and (cchh) is a 4-byte track address. The control unit saves the seek address to do a cache directory sear...

  • Page 87

    Seek head description: the seek head command transfers the track address to the control unit. The format of the seek address parameter is (00cchh), where (00) is 2 bytes of binary zeros and (cchh) is a 4-byte track address. The (hh) part of the track address replaces the (hh) value in the current tr...

  • Page 88

    Recalibrate description: the recalibrate command causes the addressed logical volume to seek to cylinder 0, head 0. This command does not cause any data transfer. Therefore, if the ccw has a nonzero byte count, the suppress length indication (sli) bit must be set to x'1' to avoid an incorrect length...

  • Page 89

    Set file mask description: the set file mask command transfers a 1-byte mask to the control unit. A valid mask byte replaces the current value of the file mask. This command resets orientation in the control unit. The format of the mask byte is the same as the define extent mask byte (see “mask byte...

  • Page 90

    Set sector description: this command resets orientation in the control unit. The set sector command transfers a 1-byte relative angular track position to the control unit. The sector number causes the logical volume to position to a specific sector before starting the next command. If the sector num...

  • Page 91

    Read sector description: this command resets track orientation in the control unit. The read sector command transfers a 1-byte sector number to the channel. The data byte represents the count-area sector position of the record that was operated on by the search or data transfer command that immediat...

  • Page 92

    Read device characteristics description: the read device characteristics command transfers 64 bytes to the channel. The information this command transfers defines the characteristics of the addressed logical volume. The eckd architecture states that the characteristics of the emulated device rather ...

  • Page 93

    Bytes 0 and 1: contains the four-digit subsystem type number as four hexadecimal digits. Table 4-20. Device characteristics data bytes contents 0–1 subsystem type 2 subsystem model number and architecture 3–4 device type 5 device model 6–9 subsystem and device facilities 10 device class code 11 devi...

  • Page 94

    Byte 2: identifies the subsystem model number and the supported architecture. The active bits show: bytes 3 and 4: contains the four-digit device type number as four hexadecimal digits. Byte 5: contains the device model and device features. Bytes 6 through 9: shows the program visible facilities in ...

  • Page 95

    Byte 10: contains an 8-bit device class code. Byte 11: contains an 8-bit device type code. Bytes 12 and 13: contains an unsigned, 16-bit binary value that shows the number of primary cylinders on the logical volume. The highest primary cylinder address on the logical volume is one less than the valu...

  • Page 96

    Bytes 17 through 19: contains an unsigned, 24-bit binary value that shows the total number of usable bytes on each track. This value is valid after the track has been formatted with a home address and a standard record zero (key length of 0 and data length of 8). This value is the maximum number of ...

  • Page 97

    Bytes 23 through 27: contains factors, f1 through f5, to be used in the track capacity calculation formulas. See the description of byte 22 above. Bytes 28 and 29: contains an unsigned, 16-bit binary value, which is the address (cc) of the first cylinder that contains alternate tracks. Bytes 30 and ...

  • Page 98

    The appropriate sector number for a locate record or a set sector command is the smallest integer equal to or less than the result of the following formula: sector = (c0 + (2 x f7) + netspace) / (f1 x f8) where: c0 is an unsigned 16-bit binary number contained in bytes 20 and 21. F7 is an unsigned 8...

  • Page 99

    Read commands the read commands cause data to transfer from the cache to the channel. The control unit is oriented to a specific record area and sends one or more areas or records to the channel. All read commands except read ipl must be preceded in the command chain by a command that specifies (or ...

  • Page 100

    Read home address description: the read home address command transfers a partial home address area to the channel. The control unit is oriented to index when the read home address command starts. A 5-byte home address transfers to the channel. These 5 bytes are the flag byte and the 4-byte track add...

  • Page 101

    Read record zero description: the read record zero command transfers the entire record zero (count and data bytes) to the channel. When in a locate record or locate record extended domain, the control unit is oriented to home address when it receives this command. Data transfer begins with the follo...

  • Page 102

    Read count, key, and data description: the read count, key, and data command transfers the entire count, key, and data areas of a record to the channel. The control unit is oriented to the next count area and sends the count, key, and data areas of that record. Chaining and restrictions: unit check ...

  • Page 103

    Ending status: the control unit presents channel end and device end status when data transfer completes. When the command completes, the control unit is oriented to the data area of the record that was read. Chapter 4. Command descriptions 4-65.

  • Page 104

    Read key and data description: the read key and data command transfers the key and data areas of a record to the channel. If the control unit is oriented to a count area when the command starts, it sends the key and data areas of the same record. If the control unit is not oriented to a count area, ...

  • Page 105

    If the data length field in the count area contains zero, the control unit does not send any data area. The command is terminated with channel end, device end, and unit exception status. Ending status: the control unit presents channel end and device end status when data transfer completes. When the...

  • Page 106

    Read data description: the read data command transfers the data area of a record to the channel. If the control unit is oriented to a count or key area when the command starts, it sends the data area of the same record. If the control unit is not oriented to a count or key area, it is oriented to th...

  • Page 107

    Ending status: the control unit presents channel end and device end status when data transfer completes. When the command completes, the control unit is oriented to the data area of the record that was read. Chapter 4. Command descriptions 4-69.

  • Page 108

    Read count description: the read count command transfers the count area of a record, other than record zero, to the channel. The control unit is oriented to the next count area and sends that count area to the channel. Chaining and restrictions: unit check status with the sense data containing comma...

  • Page 109

    Read multiple count, key, and data description: the read multiple count, key, and data command transfers the next record (excluding r0) and all remaining records on the track to the channel. The byte count value should be at least as large as the maximum size record that can be written on the track....

  • Page 110

    Read track description: the read track command transfers multiple records to the channel. This command is primarily for dump/restore utility programs. It sends all records from a logical volume track without first having to determine the number of records on the track and their formats. After readin...

  • Page 111

    Domain of a read tracks operation, the ‘next’ track is the track that is next in sequence. If the data area length is zero, the control unit neither sends the data area nor reports an exception status. The read operation continues with the next count area, if any, on the track. Ending status: the co...

  • Page 112

    Read track data hex code data address count a6 the main storage location of the first byte to be read. Number of bytes to be read. Description: read track data transfers multiple records from a track to the channel. Read track data is valid only in the domain of a locate record command that specifie...

  • Page 113

    Reached. If no user data area is detected on the next track, execution is terminated with status that includes unit check (no record found). Within the domain of a read tracks operation, the “next” track is the track that is next in sequence. Within the domain of a read trackset operation the “next”...

  • Page 114

    Read ipl description: the read initial program load (ipl) command causes the addressed logical volume to seek to cylinder 0, head 0. The logical volume then reads the data area of the record following r0. The read ipl command sends the data field of the first record following record zero on track 00...

  • Page 115

    Search commands the search commands cause the control unit to be oriented to a specific record area on the track before the control unit processes a read or write command. The control unit gets the search argument from the channel and then reads a record area from the logical volume or cache. It the...

  • Page 116

    Search home address equal description: the search home address equal command causes the control unit to request 4 bytes of home address data from the channel and compare that data to the 4 bytes of home address data read from cache or the logical volume. If the channel sends fewer than 4 bytes, the ...

  • Page 117

    Search id equal description: the search identifier (id) equal command causes the control unit to request a 5-byte record id (cchhr) from the channel and compare that data with the 5-byte count area id read from cache. If the channel sends fewer than 5 bytes, the control unit compares the data receiv...

  • Page 118

    Search id high description: the search id high command operates the same as a search id equal command, except that status modifier is included in ending status when the record id compares high to the search argument. Hex code data address count 51 multitrack d1 the main storage location for the reco...

  • Page 119

    Search key equal description: the search key equal command causes the control unit to compare a key data area received from the channel with the key data area read from cache. The kl field in the count area defines the length of the key area. If the kl field is zero, no data is transferred from the ...

  • Page 120

    Search key high description: the search key high command operates the same as a search key equal command except that status modifier is included in ending status when the key area compares high to the search argument. Hex code data address count 49 multitrack c9 the main storage location for the key...

  • Page 121

    Write commands that do not operate in a locate record or locate record extended domain must be preceded by a search command that compares equal on all bytes of the search field. The locate record orient operation code does not satisfy the search prerequisite for a write command. No write command is ...

  • Page 122

    Write home address hex code data address count 19 the main storage location of the home address. 5 description: the write home address command simulates updating the home address record on the track, by verifying a 5-byte section of the home address record. This 5-byte section consists of a flag byt...

  • Page 123

    Write record zero hex code data address count 15 the main storage location of the r0 count, key and data bytes. 16 description: the write record zero command formats a new record zero on the track. Record zero is always the first record following the home address area. The first 8 bytes of the recor...

  • Page 124

    Write count, key, and data hex code data address count 1d the main storage location of the count (8 bytes), key and data bytes. The number of bytes to write. Description: the write count, key, and data (ckd) command formats a new record on the track. The first 8 bytes of the record are the count are...

  • Page 125

    Ending status: if additional commands are expected within the domain of a locate, device end will be presented with channel end. Otherwise channel end will be presented when the data transfer is complete to the cache. When the command completes, the control unit is oriented to the data area of the r...

  • Page 126

    Write count, key, and data next track hex code data address count 9d the main storage location of the count (8 bytes), key and data bytes. The number of bytes to write. Description: the write count, key, and data (ckd) next track command formats the first user data record on the next track. The firs...

  • Page 127

    Ending status: if additional commands are expected within the domain of a locate, device end will be presented with channel end. Otherwise channel end will be presented when the data transfer is complete to the cache. When the command is complete, the control unit is oriented to the data area of the...

  • Page 128

    Erase hex code data address count 11 the main storage location of the count, key, and data bytes. The number of count, key, and data bytes. Description: the erase command erases one or more user data records from the track. The remaining portion of the track is erased starting from the end of the re...

  • Page 129

    Write update key and data hex code data address count 8d the main storage location of the data. The number of bytes to write. Description: the write update key and data command writes the key and data areas of an existing record. If the key length (kl) is zero, this command operates the same as a wr...

  • Page 130

    Chaining and restrictions: unit check status with the sense data containing command reject with format 0, message 2 occurs if: this command is not in a locate record or locate record extended domain. This command is in a locate record or locate record extended domain that did not specify a write dat...

  • Page 131

    Write key and data hex code data address count 0d the main storage location of the data. The number of bytes to write. Description: the write key and data command updates the key and data areas of an existing record. If the key length (kl) is zero, this command operates the same as a write data comm...

  • Page 132

    Sense data containing command reject with format 0, message 2 occurs if this command is in the domain of a locate record that specifies any other operation. If the record has a data area length of zero and ckd conversion mode is specified (define extent global attributes byte, bit 2 is a ‘1’), no da...

  • Page 133

    Write update data hex code data address count 85 the main storage location of the data. The number of bytes to write. Description: the write update data command updates the data area of an existing record. The control unit writes data into the data area of the record. The amount of data is defined i...

  • Page 134

    Status is presented when the write operation is completed at the logical volume. When the command completes, the control unit is oriented to the data area of the record just updated. 4-96 internal disk subsystem reference guide.

  • Page 135

    Write data hex code data address count 05 the main storage location of the data. The number of bytes to write. Description: the write data command updates the data area of an existing record. Processing depends on whether this command is inside of or outside a locate record or locate record extended...

  • Page 136

    Outside the domain of a locate record the control unit is oriented to a count or key area when the command starts (as the result of the preceding search command). It then writes the following data area. The data area length is defined in the data length (dl) field. If the channel sends fewer than (d...

  • Page 137

    Write full track hex code data address count 95 the main storage location of the first parameter byte. Number of bytes to be transferred. Description: the write full track command transfers data from the channel and writes the data on the track. Write full track is valid only in the domain of a loca...

  • Page 138

    Channel to make this determination, execution is terminated with status that includes unit check (command reject, format 0, message 3, ccw byte count less than required). If it is the pseudo count field then the track image transfer is complete, the remainder of the track is erased. 2. Otherwise the...

  • Page 139

    ┌─────┬────┬─────┬────┬─────┬────┬┬─────┬────┬──────┐ │ r │ r │ r1 │ r1 │ r2 │ r2 │ │ rn │ rn │pseudo│ │count│data│count│key+│count│key+│ │count│key+│count │ │ │ │ │data│ │data│ │ │data│ │ └─────┴────┴─────┴────┴─────┴────┴┴─────┴────┴──────┘ │ │ │ │ │ │ │ │8 bytes + │8 bytes + │8 bytes + │ │8 bytes...

  • Page 140

    Write track data hex code data address count a5 the main storage location of the first data byte to be written. Number of data bytes to be written. Description: the write track data command transfers multiple records from the channel to a track. Write track data is valid only in the domain of a loca...

  • Page 141

    Ending status: channel end status is presented when data transfer from the channel is completed. If there are further tracks to be processed (more ‘1’ bits in the extended parameter), then device end status is included with channel end. After data transfer for the last track, device end status is pr...

  • Page 142

    Sense commands sense commands send error information and specific device information to the channel. Sense description: the sense command sends 32 bytes of sense data from the control unit to the channel. See chapter 5: "24-byte compatibility sense data" on page 5-1 and chapter 6: "eckd 32-byte sens...

  • Page 143

    Sense id description: the sense id command sends 20 bytes of data describing the type and model number of the subsystem and logical volume to the channel. If the logical volume is not busy, the command will operate even if the logical volume is in the not-ready state (or absent). If the device has n...

  • Page 144

    Table 4-26 (page 2 of 2). Data sent to channel by sense id command bytes description value 16–19 command interface word (ciw) for read-node-identifier x'423e0040' 4-106 internal disk subsystem reference guide.

  • Page 145

    Miscellaneous commands no-operation description: the no-operation command does not cause any data transfer. Therefore, if a nonzero byte count is specified in the command, the sli bit must be set to ‘1’ to avoid an incorrect length exception. Also, a program check occurs on some systems if the count...

  • Page 146

    Read and reset buffered log description: the read and reset buffered log command sends 32 bytes of data to the channel. This command is implemented for compatibility purposes only. A real 3990 control unit with 3380 and 3390 devices would report overflows of usage counters and other device usage sta...

  • Page 147

    Table 4-27. Buffer log data for 3390 models byte contents 0–1 reserved, set to zeros. 2 emulated control unit type code, set to the value in byte 42 of read device characteristics data. Set to x'05'. 3 reserved, set to zero. 4 unit address of the logical volume. 5 device type code of the real device...

  • Page 148

    Table 4-28. Buffer log data for 3380 models byte contents 0–1 unused, set to zeros. 2 environmental data present. Set to x'10' 3 reserved, set to zero. 4 unit address of the logical volume. 5–6 unused, set to zeros. 7 format/message byte. Set to x'60' 8–11 reserved for bytes read. Set to zeros. 12–1...

  • Page 149

    Read configuration data description: the read configuration data command sends 256 bytes of configuration data to the channel. This command shows the host how the internal disk subsystem is configured. Also included are part numbers and manufacturer ids that the customer may use for inventory contro...

  • Page 150

    Configuration data the configuration data descriptions follow. Each separate element contains 32 bytes. Table 4-30 (page 1 of 5). Internal disk configuration data byte(s)(1) description bytes 0-31:(2) ned1; contains the ned for the volume 0 (0) flags bit description 0-1 field identifier; set to b'11...

  • Page 151

    Table 4-30 (page 2 of 5). Internal disk configuration data byte(s)(1) description 32 (0) flags bit description 0-1 field identifier; set to b'11' to indicate a ned 2 token indicator; set to b'0' 3-4 serial-number indicator; set to b'00' to indicate that this serial-number is identical to the serial-...

  • Page 152

    Table 4-30 (page 3 of 5). Internal disk configuration data byte(s)(1) description 67 (3) level bit description 0-6 reserved; set to zeros 7 level; set to b'0' to indicate that this ned has a hierarchical relationship to the next ned in this configuration record 68-73 (4-9) type number (ebcdic); set ...

  • Page 153

    Table 4-30 (page 4 of 5). Internal disk configuration data byte(s)(1) description 114-117 (18-21) sequence number high (ebcdic); this portion of the sequence number field is used to hold the control unit number. These characters are in ebcdic notation, representing four hexadecimal characters. This ...

  • Page 154

    Table 4-30 (page 5 of 5). Internal disk configuration data byte(s)(1) description 234 (10) not used; set to x'00' 235(5) (11) channel connection address; this is the unit address for this volume. 236(5) (12) physical device id; this is the same as channel connection address (byte 11 above). 237 (13)...

  • Page 155

    Restore description: the restore command does not cause any data transfer. Therefore, if the ccw has a nonzero byte count, the suppress length indication (sli) bit must be set to ‘1’ to avoid an incorrect length exception. Note: do not use this command when writing new applications. This command is ...

  • Page 156

    Path control commands path control commands are primarily intended for environments where the disk data sets are shared among application programs. The sharing may be across two applications that are each running under different operating systems both of which are controlled by vm, or data sets may ...

  • Page 157

    Chaining and restrictions: unit check status with the sense data containing command reject with format 0, message 2 occurs if: this command is received in the domain of a locate record or locate record extended command. This command was not the first command in the channel program or chained directl...

  • Page 158

    Device release description: the device release command transfers 32 bytes of sense data to the channel. If the logical volume is busy or reserved to another lpar, the device release command is i/o queued. Internal disk queues ccw chains using a fifo queue discipline. While the chain is queued, the c...

  • Page 159

    Unconditional reserve description: the unconditional reserve command transfers 32 bytes of sense data to the channel. This command terminates the addressed logical volume’s allegiance or reservation to a channel path or path group, and establishes a device reservation to the channel path group of th...

  • Page 160

    The channel had a reservation for this logical volume and did not have an active command chain. (unit check will be pending to the next start i/o.) the channel had a contingent allegiance for the logical volume, and sent a sense command for that logical volume after the unconditional reserve command...

  • Page 161

    Reset allegiance description: the reset allegiance command transfer 32 bytes of data to the channel. This command terminates a logical volume’s allegiance to a channel path or path group. It operates if the logical volume is either busy or not ready. However, it will not reset any allegiances if the...

  • Page 162

    Chaining and restrictions: unit check status with the sense data containing command reject with format 0, message 2 occurs if: this command is received in the domain of a locate record or locate record extended command. This command is not the first command in the channel program or chained directly...

  • Page 163

    Set path group id description: the set path group id command transfers 12 bytes to the control unit from the channel. The first byte is a function control byte and the following 11 bytes are the channel path group id. The channel path group identifier identifies the system control program governing ...

  • Page 164

    Single-path mode: i/o operations must be maintained for the specific channel path over which the start i/o was initiated. Grouping exists only for device reservation, not reconnection. Internal disk supports only single-path mode. Multipath mode: i/o operations started on one channel path may comple...

  • Page 165

    When a mode bit is established for a path group, a following mode bit is not the same. The function control byte bits 1 and 2 are ‘11’. Bits 3 through 7 of the function control byte are not zeros. Ending status: the control unit presents channel end and device end after it validates the parameters. ...

  • Page 166

    Sense path group id description: the sense path group id command transfers 12 bytes of path group data to the channel. The 12 data bytes are a portion of the addressed logical volume’s path group data associated with the channel path where this command is executed. Byte 0 contains the path state byt...

  • Page 167

    Chaining and restrictions: unit check status with the sense data containing command reject with format 0, message 2 occurs if this command: is in the domain of a locate record or locate record extended. Is not the only command in the channel program or chained directly from a suspend multipath recon...

  • Page 168

    Suspend multipath reconnection description: the suspend multipath reconnection command does not cause any data transfer. Therefore, if a nonzero byte count is specified in the command, the sli bit must be set to ‘1’ to avoid an incorrect length exception. This command is for recovery operations in m...

  • Page 169

    Chaining and restrictions: the perform subsystem function command cannot occur in the domain of a locate record or locate record extended. If the command is in a locate record or locate record extended domain, it is rejected with unit check status. The sense data contains command reject with format ...

  • Page 170

    Valid parameters include the following: the flag byte and bytes 2–5 must be zero. Parameter byte 6 must contain a valid suborder. The only valid suborders are x'00' and x'01' and x'03'. Byte 7 must be zero. Bytes 8 through 11 must be zero unless the suborder in byte 6 is x'03', in which case they co...

  • Page 171

    Are called "global" commands. Set special intercept condition causes global commands to be rejected. Internal disk has only two of the global commands that this order is designed to intercept. When the host operating system has conditioned a logical volume with this order, the commands that internal...

  • Page 172

    Sense subsystem status (page 4-136) the exchange of information about host and subsystem support of commands and op codes does not alter the execution of the commands and op codes. If a host denies support of the prefix command (byte 6, bit 0 set to zero) and later issues the prefix command, interna...

  • Page 173

    No validity checks are made on parameter bytes 2–65. The subsystem will ignore any bits set that it does not understand. This information is retained by the subsystem as long as the path group is valid. Set interface identifier x'b0' the set interface identifier order can set an interface identifier...

  • Page 174

    Rejected with unit check status. The sense data contains command reject with format 0, message 2. Ending status: when the parameters have been validated, channel end and device end are presented. Sense subsystem status the sense subsystem status command causes the internal disk to send 40 bytes of i...

  • Page 175

    The byte descriptions follow: byte 1, logical volume address.: the address the channel program is using. Byte 2, number of logical volumes with statistics: the number of logical volumes in the subsystem accessible from the channel minus one. The value of byte 2 ranges from x'00' to x'ff' to describe...

  • Page 176

    Byte 14 – 17, available cache capacity in bytes: this field contains the same value as that in bytes 10 – 13. Byte 26, internal disk fast write status: bits 0-1 are reserved and set to zeros. Bits 2 and 3 are used to reflect the state of internal disk raid fast write (idrfw) for the logical volume. ...

  • Page 177

    Read subsystem data the read subsystem data command causes the internal disk to send the data requested by the perform subsystem function command, from which the read subsystem data command is chained. Description: see the description of the specific read subsystem data type. Chaining and restrictio...

  • Page 178

    Storage subsystem status prepare for read subsystem data with byte 6 equal to x'00', a 16-byte record is returned which consists of the first byte set to '48'x and the remaining bytes set to zeros. The 3990-2 returns storage path status in these bytes. Table 4-40 shows the contents of the record ret...

  • Page 179

    Table 4-41 (page 2 of 2). Performance statistics bytes description 32–35 write sequential i/o operation hits. The number of sequential operations that had at least one write command but did not cause any portion of a track to be staged from the disk subsystem to cache. 36–51 reserved, for cache fast...

  • Page 180

    Node descriptor record if the preceding perform subsystem function command specified a set-interface-identifier order (x'b0') then internal disk returns 32 bytes of data. The data returned has the following format: table 4-43 (page 1 of 2). Node descriptor format byte description bytes 0-31 contain ...

  • Page 181

    Table 4-43 (page 2 of 2). Node descriptor format byte description 32 flags bit description 0-2 node qualifier contents, set to b'000' meaning: the node-qualifier field contains a list of interface ids. Because internal disk has only one chpid, the list has a single entry that points back to the only...

  • Page 182

    4-144 internal disk subsystem reference guide.

  • Page 183

    Chapter 5. 24-byte compatibility sense data the internal disk sense information sends 32 bytes to the host for the following reasons: identifies the cause of the last unit check status presented for the logical volume on this channel interface. Provides data that may be required for system error rec...

  • Page 184

    24-byte sense data summary the internal disk sense data is presented to the cpu as 24-byte compatibility sense when byte 27, bit 0, is set to ‘1’. Table 5-2 on page 5-3 contains a summary of the 24-byte compatibility sense data. Sense data formats the first 7 bytes (bytes 0 through 6) and last 8 byt...

  • Page 185

    Table 5-2. Sense information summary–24-byte compatibility byte bits meaning page 0 0 1 2 3 4 5–6 7 command reject intervention required reserved, set to zero equipment check data check reserved, set to zero incomplete domain 5-4 1 0 1 2 3 4 5 6 7 permanent error invalid track format end-of-cylinder...

  • Page 186

    24-byte sense data description when byte 27, bit 0 is ‘1’, the sense data uses 24-byte compatibility mode. Bytes 0 through 7 and 24 through 31 are common to all formats in 24-byte compatibility sense. Bytes 8 through 23 depend on the format in bits 0 through 3 of byte 7. Sense bytes 0, 1, and 2 the ...

  • Page 187

    Note: when an incomplete domain exception occurs, it may prevent the unit check for a subsequent exception from being presented with ending status. Byte 1, bit 0–permanent error: a modifier bit that overrides any other possible bit settings. When set to ‘1’, this bit shows that host program error re...

  • Page 188

    3. In the domain of a locate record or locate record extended, either r0 or a user record could not be found after a head or cylinder switch. Byte 1, bit 5–file protected: can be caused by a programming error or an expected programming condition. It occurs for one of the following: 1. A seek operati...

  • Page 189

    Bits 2-7 contain the logical volume address. Sense byte 5–low cylinder address represents the least significant bits of the last cylinder argument acted upon at this logical volume. For a 3390/9 logical volume, the cylinder and head address can not be contained in two bytes. Therefore, the cylinder ...

  • Page 190

    1d a long busy condition exists in the subsystem or logical volume. Environmental-data present (byte 2, bit 3) is a ‘1’ and byte 7 contains x'f0'. 70 a command was rejected because the interface was disabled for specific commands by the set special intercept condition order of the perform subsystem ...

  • Page 191

    Table 5-5 (page 2 of 2). Sense bytes 29 through 31 byte bits content 30 0 1 2 3 4 5 6 7 cylinder 128 cylinder 64 cylinder 32 cylinder 16 cylinder 8 cylinder 4 cylinder 2 cylinder 1 31 0–3 4 5 6 7 reserved head 8 head 4 head 2 head 1 chapter 5. 24-byte compatibility sense data 5-9.

  • Page 192

    24-byte compatibility sense data formats format 0–program or system checks this format reports program or system check conditions. Sense bytes 8 through 23 for format 0 format 0 sense bytes 8 through 23 contain the following: table 5-6. Sense bytes 8–23 – format 0 bytes definition 8 the reason code ...

  • Page 193

    Message 9–e: reserved. Message f–status not as required: occurs when one of the following byte 8 reason codes is specified. Code reason 00 no message. 01–28 reserved. 29 trying to do a perform subsystem function command with the set special intercept condition order on an interface on which no path ...

  • Page 194

    Microcode detected data discrepancy, message 3 intervention required, message 9, due to: 1. Not physically attached to internal disk 2. Not available for use because the drive motor is off or is powering on. 3. The hdd has been disabled via the service element. 4. The ssa adapter card has been remov...

  • Page 195

    Table 5-8. Format 1 messages byte 7 bits 4-7 value format 1 message symptom code 0 device fails to respond 'e000'x 1 predictive failure analysis 'e101'x 2 reserved – 3 microcode detected data discrepancy. 'e300'x 4-8 reserved – 9 intervention required 'e210'x a-f reserved – format 2–internal disk co...

  • Page 196

    Sense bytes 8 through 23 for format 3 format 3 sense bytes 8 through 23 contain the following: table 5-11. Sense bytes 8–23 – format 3 bytes definition 8–10 reserved, set to zeros 11–12 hardware level id for field-replaceable unit. 13–14 reserved, set to zeros. 15–19 the plant of manufacture and ser...

  • Page 197

    Format 5 to format e – not used format f–subsystem checks two messages are used under format f to report: long busy condition using message 0. (defective sector reassignment is an example.) inability to determine the internal disk raid fast write status of a logical volume using message c. Sense byt...

  • Page 198

    5-16 internal disk subsystem reference guide.

  • Page 199

    Chapter 6. Eckd 32-byte sense data the internal disk sends 32 bytes of sense information to the host for the following reasons: provides data that may be required for system error recovery. Provides information that aids the service representative in diagnosing and isolating subsystem malfunctions, ...

  • Page 200

    Eckd 32-byte sense data summary the internal disk sense data is presented to the cpu as eckd 32-byte sense when byte 27, bit 0, is ‘0’. Byte 6 contains the content and format of the other bytes. The contents of bytes 7 through 14 are dependent on the exception class in bits 0–3 of byte 22 and the fo...

  • Page 201

    Table 6-2. Eckd 32-byte sense data summary byte bits meaning page 0 0 1 2 3 4 5–6 7 command reject intervention required reserved equipment check data check reserved incomplete domain 6-4 1 0 1 2 3 4 5 6 7 permanent error invalid track format reserved, set to zero message to operator reserved, set t...

  • Page 202

    Eckd 32-byte sense data description when byte 27, bit 0 is ‘0’, the sense data uses 32-byte mode. Sense byte 0 byte 0, bit 0–command reject: occurs for any of the following conditions. (sense byte 7 identifies the error condition in more specific terms.) 1. A command, or a sequence of commands is no...

  • Page 203

    Byte 1, bit 1–invalid track format: indicates that an invalid track condition is set for the following: 1. An update write operation is attempted on a record whose size differs from the record size parameter (tlf). 2. An attempt is made to exceed the track capacity of the track. 3. A previously inva...

  • Page 204

    Sense byte 5–device type code for exception classes 4 and e this byte contains a code that defines the device type and model. For 9gb ultrastar2 xp drives on the internal disk subsystem, the code is x'3b'. This code is the same as byte 57 of the data returned by the read device characteristics comma...

  • Page 205

    Sense byte 24–logging and message control the definitions in byte 24 are: bits 4 and 5–logging action: these bits show the error recovery procedure (erp) how to log this error. The definitions of these bits are: the logging action is to be performed in addition to any other program action that is sp...

  • Page 206

    Message action. A program is not required to interpret additional sense data to determine if an operator message is required. When bits 6–7 are ‘10’, a message to the operator is to be sent once within a retry sequence on a path whether or not the retry action is successful. When bits 6–7 are ‘11’, ...

  • Page 207

    Sense bytes 29 and 30–cylinder address if byte 6, bit 2 is ‘1’, these bytes contain a valid value. For exception classes 4 and e, these bytes contain the cylinder address of the most recent seek argument accepted from the channel or executed by an internal operation. Table 6-9. Bytes 29 and 30 – 32-...

  • Page 208

    Sense data dependent on exception class bits 0–3 of byte 22 contain the exception class for eckd 32-byte sense data. Bits 4–7 of byte 22 describe the category of exceptions. Byte 6, bits 4–7, define the format of the data in bytes 7–19 for each exception class. Exception class 0–i/o program exceptio...

  • Page 209

    Exception class 0–format 4: format 4 presents sense data associated with the i/o program exceptions caused by a locate record or locate record extended domain that contained fewer data transfer commands than were required to transfer the number of records or tracks specified by the count field. This...

  • Page 210

    Table 6-12 (page 2 of 2). Exception class 0 – format 4. (byte 6, bits 4–7 set to x'4'.) byte contents 7 operation byte (byte 0) contains the operation byte (byte 0) of the locate record or locate record extended command that established the domain. 8 contains the extended operation code from byte 17...

  • Page 211

    Table 6-13. Exception class 4 – format 1. (byte 6, bits 4–7 set to x'1'.) byte contents 7–19 unused, set to zero 20–21 ssid 22 exception class and category. Set to '44'x. 23 exception condition. Set to 'c0'x. Indicating permanent error, not recovered. 24 bits description 0-3 reserved 4-5 '10' log on...

  • Page 212

    Category 2–intervention required: shows that the addressed logical volume is in one of the following states: 1. Not physically attached to internal disk 2. Not available for use because the drive motor is off or is powering on. 3. The hdd has been disabled via the service element. 4. The ssa adapter...

  • Page 213

    Table 6-17. Exception class e – category 3, bytes 6 – 25 byte contents 6 x'e1' content and format 7–19 reserved, set to zero. 20–21 subsystem id (ssid) 22 x'e3' exception class and type 23 set to zero. 24 x'0a' log and message once 25 x'83' retry ccw chain and do 255 retries. Chapter 6. Eckd 32-byte...

  • Page 214

    6-16 internal disk subsystem reference guide.

  • Page 215

    Chapter 7. Error recovery procedures the internal disk provides 24-byte compatibility sense data and 32-byte sense data. The 24-byte compatibility sense indicates the error recovery and logging actions in bytes 0–2. The 32-byte sense indicates these actions in bytes 24 and 25. “erps for 24-byte comp...

  • Page 216

    Table 7-1 (page 2 of 2). Error conditions byte bit sense data condition act log 1 2 end-of-cylinder end-of-cylinder during a multitrack operation outside the domain of a locate record or locate record extended. 7 no 2 3 environmental-data present (with format 0, message f) attention was presented on...

  • Page 217

    Action 5: 1. If the chain has not been retried 10 times, repeat the command chain. 2. Do action 1. Action 6 not used. Action 7: 1. Increment the cylinder address of the seek argument in sense bytes 5 and 6 (or 29 and 30) by one. Reset the head address to zero. If the incremented cylinder address is ...

  • Page 218

    D. Continue the operation with the following ccw chain: 3. If the interrupted ccw was a locate record command: a. Determine if the seek address is in the user’s extents. If no seek address exists, do action 2. B. Continue the operation with the following ccw chain: 4. This is either a multitrack ccw...

  • Page 219

    Define extent the user’s new extent is compatible with the seek argument from steps 4a and 4b. The other parameters are the same as the original parameters. Locate record bytes contents 0 same as in original locate record command, except: if sense byte 3 equals ‘01’ and the original locate record co...

  • Page 220

    Erp actions error recovery program actions are indicated by the contents of sense byte 25. Byte 25 contains either a single or a compound program action code. If bit 0 of byte 25 is ‘0’, bits 1–7 contain a single program action code that identifies a recovery action for one specific condition. See “...

  • Page 221

    Program must obtain this data from the original command within processor storage since this information is not available in the sense data. Error recovery programs may be unable to find the original locate record extended command when the command follows a part of a channel program which contains co...

  • Page 222

    Action 15: a multitrack data transfer operation within the domain of a locate record or locate record extended command has attempted to switch to a new track and the next track is outside the defined extent. Sense bytes 29–31 contain the address ‘cch’ of the last track operated upon. Recovery steps:...

  • Page 223

    The original locate record cannot be found, then exit with a permanent error indicator. Sense bytes 29–31 contain the address ‘cch’ of the track where the channel program restarts. Recovery steps: 1. Determine the address of the ccw to be restarted using the count of the completed operations and the...

  • Page 224

    Message code (byte 28) defines. The operator message control can specify messages to be sent unconditionally, sent once during an erp sequence, or sent if the procedure is unsuccessful. (see “sense byte 24–logging and message control” on page 6-7.) 7-10 internal disk subsystem reference guide.

  • Page 225: Appendix A.

    Appendix a. Device characteristics device characteristics information is shown in the following device type tables. This data is returned in response to a read device characteristics command. See “read device characteristics” on page 4-54. 3390 logical volumes table a-1 on page a-2 "3390 model 1/2/3...

  • Page 226

    Table a-1 (page 1 of 3). 3390 model 1/2/3/9 device characteristics bytes description emulated 3390 0–1 subsystem type x‘3990’ 2 subsystem model and architecture x‘e2’ 3–4 device type x‘3390’ 5 device model 3390-1 3390-2 3390-3 3390-9 x‘02’ x‘06’ x‘oa’ x‘0c’ 6 subsystem and device facilities bits def...

  • Page 227

    Table a-1 (page 2 of 3). 3390 model 1/2/3/9 device characteristics bytes description emulated 3390 11 device type code 3390-1 3390-2 3390-3 3390-9 x‘26’ x‘27’ x‘24’ x‘32’ 12–13 primary cylinders 3390-1 1113 cyl 3390-2 2226 cyl 3390-3 3339 cyl 3390-9 10017 cyl x‘0459’ x‘08b2’ x‘0d0b’ x‘2721’ 14–15 tr...

  • Page 228

    Table a-1 (page 3 of 3). 3390 model 1/2/3/9 device characteristics bytes description emulated 3390 54 device and control unit features bits definition 0 reserved 1 raid device 2 reserved 3 transparent subsystem cache 4–7 reserved b‘0’ b‘1’ b‘0’ b‘1’ b‘0000’ 55 reserved x‘00’ 56 real control unit typ...

  • Page 229

    Table a-3. 3390 models 1/2/3/9 access authorization address range description emulated 3390 normal access * low high 3390-1 1113 cyl 3390-2 2226 cyl 3390-3 3339 cyl 3390-9 10017 cyl x‘00000000’ x‘0458000e’ x‘08b1000e’ x‘0d0a000e’ x‘2720000e’ device support * low high 3390-1 1113 cyl 3390-2 2226 cyl ...

  • Page 230

    Table a-4 (page 1 of 3). 3380 models j/e/k device characteristics bytes description emulated 3380 0–1 subsystem type x‘3990’ 2 subsystem model and architecture x‘e2’ 3–4 device type x‘3380’ 5 device model 3380-j 3380-e 3380-k x‘16’ x‘0a’ x‘1e’ 6 subsystem and device facilities bits definition 0 mult...

  • Page 231

    Table a-4 (page 2 of 3). 3380 models j/e/k device characteristics bytes description emulated 3380 12–13 primary cylinders 3380-j 885 cyl 3380-e 1770 cyl 3380-k 2655 cyl x‘0375’ x‘06ea’ x‘0a5f’ 14–15 tracks per cylinder x‘000f’ 16 number of sectors x‘de’ 17–19 track length x‘00bb60’ 20–21 length of h...

  • Page 232

    Table a-4 (page 3 of 3). 3380 models j/e/k device characteristics bytes description emulated 3380 58–63 reserved zeros note: * internal disk supports this function, however, this bit will only be set if the system has also indicated support of the function in a prior command. See "set system charact...

  • Page 233

    Table a-6. 3380 models j/e/k volumes access authorization address range description emulated 3380 normal access * low high 3380-j 885 cyl 3380-e 1770 cyl 3380-k 2655 cyl x‘00000000’ x‘0374000e’ x‘06e9000e’ x‘0a5e000e’ device support * low high 3380-j 885 cyl 3380-e 1770 cyl 3380-k 2655 cyl x‘0000000...

  • Page 234

    A-10 internal disk subsystem reference guide.

  • Page 235

    Appendix b. Caching algorithms caching algorithm definitions when access to data is required the control unit first searches the cache directory for the track image. Track images in cache may take two forms. If the entire track from index to end of data is present the image is a full track image. A ...

  • Page 236

    If the reference is to the front of a partial track image, stage the track from the point of reference up to the start of the partial track image and treat the operation as a hit. If the reference is to ha or r0 of the track, stage the track from index to the end of data. Execute the channel program...

  • Page 237

    – write miss execute the channel program using the cache and write the data to the disk subsystem. Invalidate copy of the track in the cache. – write hit invalidate copy of the track in the cache. Read the data into the cache and execute the channel program using this cache image and write the data ...

  • Page 238

    B-4 internal disk subsystem reference guide.

  • Page 239: Appendix C. Warranties

    Appendix c. Warranties statement of limited warranty the warranties provided by ibm in this statement of limited warranty apply only to machines you originally purchase for your use, and not for resale, from ibm or an ibm authorized reseller. The term “machine” means an ibm machine, its features, co...

  • Page 240

    Replacement parts assume the remaining warranty of the parts they replace. If a machine does not function as warranted during the warranty period, ibm or your reseller will repair or replace it (with a machine that is at least functionally equivalent) without charge. If ibm or your reseller is unabl...

  • Page 241

    In canada, warranties include both warranties and conditions. Some jurisdictions do not allow limitations on how long an implied warranty lasts, so the above limitation may not apply to you. Limitation of liability circumstances may arise where, because of a default on ibm's part (including fundamen...

  • Page 242

    Ibm agreement for licensed internal code you accept the terms of this agreement by your initial use of a machine that contains ibm licensed internal code (called “code”). These terms apply to code used by certain machines ibm or your reseller specifies (called “specific machines”). International bus...

  • Page 243

    4. Lease the code or any copy of it. Appendix c. Warranties c-5.

  • Page 244

    C-6 internal disk subsystem reference guide.

  • Page 245: Appendix D. Notices

    Appendix d. Notices this information was developed for products and services offered in the u.S.A. Ibm may not offer the products, services, or features discussed in this document in other countries. Consult your local ibm representative for information on the products and services currently availab...

  • Page 246

    If you are viewing this information softcopy, the photographs and color illustrations may not appear. Trademarks the following terms are trademarks of the international business machines corporation in the united states, or other countries, or both: other company, product, and service names may be t...

  • Page 247

    Department of communications. Operation in a residential area may cause unacceptable interference to radio and tv reception requiring the owner or operator to take whatever steps are necessary to correct the interference. Avis de conformlté aux normes du ministère des communications du canada cet éq...

  • Page 248

    D-4 internal disk subsystem reference guide.

  • Page 249: List of Abbreviations

    List of abbreviations this list contains definitions for acronyms and abbreviations used in the various books in the storage subsystem library. Some terms are more specifically defined in the glossary. Acb access method control block acs automatic class selection acds active control data set ascii a...

  • Page 250

    Map maintenance analysis procedure mb megabyte mru most recently used mvscp mvs configuration program os operating system pam partitioned access method pfk program function keyboard plpa pageable link pack area ptf program temporary fix r0 record zero raid redundant array of independent devices rmf ...

  • Page 251: Glossary

    Glossary this glossary contains direct access storage subsystem terms used in this book. This glossary includes definitions from the ibm dictionary of computing, mcgraw-hill, new york (1994) a actuator. A set of access arms and their attached read/write heads that moves as an independent component w...

  • Page 252

    Drawer. A removable direct-access storage unit which is installed in a frame. A drawer consists of disk drive modules. E effective data rate. The rate at which data can be moved over a number of i/o operations. The effective data rate depends on several factors such as the distance between a channel...

  • Page 253

    P predictive failure analysis. Failure analysis techniques that predict errors before they actually occur. Primary track. On a direct access storage device, the original track on which data is stored. See also alternate track. R release. A facility that allows other host systems to communicate with ...

  • Page 254

    X-6 internal disk subsystem reference guide.

  • Page 255: Index

    Index numerics 24-byte compatibility sense data description 5-1 format 5-10 summary 5-2 32-byte sense data see eckd 32-byte sense data 3390 device characteristics, 3390 a-1 a abbreviations x-1 access authorization 4-10 device support 4-10 diagnostic 4-10 mask byte, define extent parameter 4-10, 4-13...

  • Page 256

    Command (continued) descriptions (continued) set path group id 4-125 set sector 4-52 suspend multipath reconnection 4-130 unconditional reserve 4-121 write ckd 4-86 write ckd next track 4-88 write data 4-97 write full track 4-99 write home address 4-84 write key and data 4-93 write record zero 4-85 ...

  • Page 257

    Exception (continued) classes, sense data 6-10 code, sense byte 6-6 conditions 4-2 extent beginning of 4-13 definition 3-3 end of address 4-13 f file mask, description 3-4 mask, in define extent command 4-8 protected, sense byte 5-6 protection 3-4 set file mask command 4-51 format 24-byte compatibil...

  • Page 258

    Logging and message control, sense byte 6-7 logging mode, cache 4-12 long busy 1-7, 5-15, 7-2 m mask byte, define extent command 4-8 description 3-4 set file mask command 4-51 media maintenance 1-6 message code, eckd 32-byte sense data 6-8 erp requirements 7-9 erp, message 1 7-5 to operator, sense b...

  • Page 259

    Read track command 4-72 read tracks, locate record operation 4-24 read trackset, locate record extended operation 4-37 read, locate record operation 4-25 recalibrate command 4-50 record accessing 3-3 end-of-file 3-4 user data 3-1 recovery procedures, error 7-1 reset notification 2-10 notification, m...

  • Page 260

    Status (continued) initial 2-4 modifier 2-2 not as required 5-11 pending 2-4 presentation 2-1 presented to commands, summary 2-5 retry 2-9 stacked 2-4 state change 2-5 storage subsystem 4-140 unit check 2-3, 2-4 unit exception 2-3, 3-4 without device end 2-10 subsystem status effect on command 2-5 p...

  • Page 261

    Readers' comments — we'd like to hear from you s/390 internal disk subsystem: reference guide publication no. Sa22-1025-00 is there anything you especially like or dislike about this book? Feel free to comment on specific errors or omissions, accuracy, organization, or completeness of this book. Ibm...

  • Page 262: Business Reply Mail

    Cut or fold along line cut or fold along line readers' comments — we'd like to hear from you sa22-1025-00 ibm fold and tape please do not staple fold and tape no postage necessary if mailed in the united states business reply mail first-class mail permit no. 40 armonk, new york postage will be paid ...

  • Page 264

    Ibm printed in the united states of america on recycled paper containing 10% recovered post-consumer fiber. Sa22-125-.