Hitachi

Hitachi Advanced Database Setup and Operation Guide

12.9.2 Column structure of table function derived table when retrieving audit trails

This section explains the column structure of table function derived tables when retrieving audit trails.

The following table shows the column structure of table function derived tables when retrieving audit trails:

Table 12‒10: Column structure of table function derived table when retrieving audit trails

No.

Column name

Column data type

Description

NOT NULL constraint

1

HADB_VERSION

CHAR(8)

The version of the HADB server that received the event.

None

2

AUDIT_TRAIL_TYPE

VARCHAR(32)

The event that caused the audit trail to be acquired.

  • EVENT (event termination)

None

3

EXEC_TIME

TIMESTAMP(6)

The time at which the event finished.

None

4

USER_NAME

VARCHAR(100)

The authorization identifier of the HADB user who executed the event.

None

5

EVENT_TYPE

VARCHAR(32)

The type of the event.

  • SYSTEM (system event)

  • AUDIT (audit event)

  • SESSION (session event)

  • PRIVILEGE (privilege management event)

  • DEFINE (definition SQL event)

  • MANIPULATE (data manipulation SQL event)

  • COMMAND (command operation event)

None

6

EVENT_SUBTYPE

VARCHAR(64)

The subtype of the event.

For details, see Table 12‒11: Event subtypes output in audit trails.

None

7

EVENT_RESULT

CHAR(10)

Whether the event was successful.

  • SUCCESS

  • FAILURE

  • OCCURRENCE#

#

This type might be output as the event result of event type DISCONNECT. For details, see (2) Notes about executing the adbcancel command in 12.10.2 Notes about audit trails output during command execution.

None

8

USED_PRIVILEGE

VARCHAR(32)

If EVENT is output in the AUDIT_TRAIL_TYPE column, NULL is output in this column.

None

9

OS_USER_NAME

VARCHAR(256)

The OS account name of the client that issued the event.

If the OS account name cannot be identified, NULL is output.

None

10

AP_NAME

VARCHAR(30)

The application identifier of the event issuer.

If a command was executed that connects to the HADB server, the command name is output as the application identifier.

If the application identifier of the event issuer cannot be identified, NULL is output.

None

11

CLIENT_IP_ADDRESS#

VARCHAR(46)

The IP address of the client that issued the event.

If the IP address of the client that issued the event cannot be identified, NULL is output.

None

12

CLIENT_PORT_NUMBER#

SMALLINT

The port number of the client that issued the event.

If the port number of the client that issued the event cannot be identified, NULL is output.

None

13

CLIENT_PROCESS_NAME

VARCHAR(255)

The process name of the client that issued the event.

If either of the following applies, NULL is output:

  • The client that issued the event is the JDBC driver

  • The process name of the client that issued the event cannot be identified

None

14

CLIENT_PROCESS_ID

SMALLINT

The process ID of the client that issued the event.

If the process ID of the client that issued the event cannot be identified, NULL is output.

None

15

HADB_HOST_NAME

VARCHAR(255)

The host name of the HADB server that received the event.

If the host name of the HADB server cannot be identified, NULL is output.

None

16

HADB_IP_ADDRESS

VARCHAR(46)

The IP address of the HADB server that received the event.

If the IP address of the HADB server cannot be identified, NULL is output.

None

17

HADB_PORT_NUMBER

SMALLINT

The port number of the HADB server that received the event.

If the port number of the HADB server cannot be identified, NULL is output.

None

18

HADB_PROCESS_ID

SMALLINT

The process ID of the HADB server that received the event.

If the process ID of the HADB server cannot be identified, NULL is output.

None

19

ADBDIR

VARCHAR(118)

The path name of the server directory specified in the environment variable ADBDIR.

If the specified path name is longer than 118 bytes, only the first 118 bytes are output.

None

20

HADB_NODE_NUMBER

SMALLINT

When using the multi-node function, the node number (1 to 4) of the HADB server that received the event is output.

If the multi-node function is not being used, NULL is output.

None

21

CONNECTION_ID

SMALLINT

The connection ID (1 to 1,024) of the connection that executed the event.

If the connection ID cannot be identified, NULL is output.

None

22

CONNECTION_NUMBER

INTEGER

The connection sequence number (1 to 4,294,967,295) of the connection that executed the event.

If the connection sequence number cannot be identified, NULL is output.

None

23

STATEMENT_HANDLE

SMALLINT

The statement handle (1 to 4,095) used when executing the event.

If the statement handle cannot be identified, NULL is output.

None

24

SQL_SERIAL_NUMBER

INTEGER

The serial number of the SQL statement (1 to 9,223,372,036,854,775,807).

If the serial number of the SQL statement cannot be identified, NULL is output.

None

25

MESSAGE_LOG_INFO

CHAR(20)

The message log information associated with the event.

If the message log information cannot be identified, NULL is output.

None

26

EXIT_STATUS

SMALLINT

If EVENT is output in the AUDIT_TRAIL_TYPE column, either of the following is output:

  • If a SQL statement was executed, SQLCODE is output.

  • If a command was executed, the return code is output.

None

27

OBJECT_TYPE

VARCHAR(32)

The type of object specified as the target of the event.

  • INDEX (index)

  • SCHEMA (schema)

  • TABLE (base table)

  • TABLE FUNCTION (table function derived table)

  • VIEW (viewed table)

If the object type cannot be identified, NULL is output.

For details about the circumstances in which audit trails are output for each object type, see Table 12‒19: Circumstances in which object types are output in audit trails.

None

28

OBJECT_OWNER_NAME

VARCHAR(100)

The authorization identifier of the owner of the object specified as the target of the event.

If the object owner cannot be identified, NULL is output.

None

29

OBJECT_SCHEMA_NAME

VARCHAR(100)

The schema name of the object specified as the target of the event.

If the schema name of the object cannot be identified, NULL is output.

None

30

OBJECT_NAME

VARCHAR(100)

The identifier of the object specified as the target of the event.

If the object identifier cannot be identified, NULL is output.

None

31

ACCESS_COUNT

INTEGER

The number of rows of the object (base table or viewed table) that were accessed by the event. The following are included in the number of rows:

  • The number of retrieved rows (including data export)

  • The number of inserted rows (including data import)

  • The number of updated rows

  • The number of deleted rows

If the number of accessed rows cannot be acquired, NULL is output.

For details about the number of rows output in relation to each operation, see Table 12‒20: Number of rows output in ACCESS_COUNT column.

None

32

PRIVILEGE_TYPE

VARCHAR(32)

The name of the privilege specified for granting or revocation by the event.

For details about privilege names, see Table 12‒12: Privilege names output in audit trails.

If the privilege name cannot be identified, NULL is output.

None

33

PRIVILEGE_USER_NAME

VARCHAR(100)

The authorization identifier of an HADB user to whom either of the following applies:

  • The authorization identifier of an HADB user whose privilege is specified for granting or revocation

  • The authorization identifier of the HADB user specified when creating, deleting or changing the information of an HADB user

If the authorization identifier of the HADB user cannot be identified, NULL is output.

None

34

SQL_SOURCE

VARCHAR(64000)

The SQL statement executed in the event.

If one SQL statement specifies multiple objects, the SQL statement is output in the audit trail for each object.

The SQL statement is truncated after the first 64,000 bytes.

If no SQL statement was executed, NULL is output.

None

35

PARAM

VARCHAR(64000)

The dynamic parameters of the SQL statement executed in the event.

If one SQL statement specifies multiple objects, the dynamic parameters are output in the audit trail for each object.

The dynamic parameters are separated by commas (,). For example:

dynamic-parameter,dynamic-parameter,dynamic-parameter,...

The list of dynamic parameters is truncated after the first 64,000 bytes.

If no dynamic parameters are specified, NULL is output.

For details about the display format of each data type of data bound to dynamic parameters, see Table 12‒13: Output format for dynamic parameters.

None

36

BEFORE_SYSTEM_INFO

VARCHAR(2000)

The system information before being changed.

This information is output when the following events are successful:

For events other than the preceding, NULL is output.

The system information before being changed is truncated after the first 2,000 bytes.

If there were no system changes, the information in this column is the system information in effect when the event was executed. This means that the BEFORE_SYSTEM_INFO column and the AFTER_SYSTEM_INFO column will contain the same information.

None

37

AFTER_SYSTEM_INFO

VARCHAR(2000)

System information after being changed.

This information is output when the following events are successful:

For events other than the preceding, NULL is output.

The system information after being changed is truncated after the first 2,000 bytes.

If there were no system changes, the information in this column is the system information in effect when the event was executed. This means that the BEFORE_SYSTEM_INFO column and the AFTER_SYSTEM_INFO column will contain the same information.

None

38

SERVER_OPERAND

VARCHAR(4000)

The values of the operands specified in the server definition when starting the HADB server.

The operands are separated by slashes (/).

The server definition information is truncated after the first 4,000 bytes.

NULL is output unless the HADB server is starting (that is unless ADBSTART is in the EVENT_SUBTYPE column).

For details about the format of the server definition information output in the SERVER_OPERAND column, see Table 12‒14: Output format of server definition.

None

39

USER_INFO_1

VARCHAR(100)

Additional information set by the HADB user.

In either of the following cases, NULL is output:

  • Additional information set by the HADB user cannot be identified

  • The HADB user has not set any additional information

None

40

USER_INFO_2

VARCHAR(100)

None

41

USER_INFO_3

VARCHAR(100)

None
#

The following information is output as the audit trail for a SQL statement executed by an HADB client on the same machine as the HADB server:

  • IP address: 127.0.0.1

  • Port number: NULL

When using the multi-node function, the information in the audit trail for a SQL statement executed by an HADB client on the same machine as the HADB server might differ depending on the node that processed the SQL statement.

The following table shows the event subtypes output in the EVENT_SUBTYPE column in Table 12‒10: Column structure of table function derived table when retrieving audit trails.

Table 12‒11: Event subtypes output in audit trails

No.

Event category

Event type

Event subtype

Event subtype value output in audit trail

1

Mandatory audit event

System event

Starting the HADB server (adbstart command)

ADBSTART

2

Stopping the HADB server (adbstop command)

ADBSTOP

3

Changing the HADB server operation mode (adbchgsrvmode command)

ADBCHGSRVMODE

4

Changing the node type (adbchgnodetype command)

ADBCHGNODETYPE

5

Starting and stopping output of SQL trace information (adbchgsqltrc command)

ADBCHGSQLTRC

6

Centrally managing client definitions (adbclientdefmang command)

ADBCLIENTDEFMANG

7

Adding and modifying DB areas (adbmodarea command)

ADBMODAREA

8

Changing a buffer (adbmodbuff command)

ADBMODBUFF

9

Managing the updated-row columnizing facility (adbcolumnize command)

ADBCOLUMNIZE

10

Audit event

Granting audit admin privilege (GRANT statement with AUDIT ADMIN specified)

GRANT

11

Granting audit viewer privilege (GRANT statement with AUDIT VIEWER specified)

12

Revoking audit admin privilege (REVOKE statement with AUDIT ADMIN specified)

REVOKE

13

Revoking audit viewer privilege (REVOKE statement with AUDIT VIEWER specified)

14

Defining audit targets (CREATE AUDIT statement)

CREATE AUDIT

15

Deleting audit target definitions (DROP AUDIT statement)

DROP AUDIT

16

Changing auditor passwords (ALTER USER statement)

ALTER USER

17

Enabling the audit trail facility (adbaudittrail command with --start option specified)

ADBAUDITTRAIL START

18

Disabling the audit trail facility (adbaudittrail command with --stop option specified)

ADBAUDITTRAIL STOP

19

Swapping the audit trail file (adbaudittrail command with --swap option specified)

ADBAUDITTRAIL SWAP

20

Referencing information related to the audit trail facility (adbaudittrail command with -d option specified)

ADBAUDITTRAIL DISPLAY

21

  • Referencing audit target definition information (retrieval from SQL_AUDITS dictionary table)

  • Retrieving data from viewed tables that depend on the SQL_AUDITS dictionary table

SELECT

22

ADBEXPORT

23

  • Using a system-defined function for audit trails (ADB_AUDITREAD function)

  • Retrieving data from a viewed table that depends on a derived table that specifies an ADB_AUDITREAD function

SELECT

24

ADBEXPORT

25

Converting audit trail files (adbconvertaudittrailfile command)

ADBCONVERTAUDITTRAILFILE

26

Optional audit event

Session event

Connecting to the HADB server

CONNECT

27

Disconnecting from the HADB server

DISCONNECT

28

Privilege management event

Granting the DBA privilege (GRANT statement with DBA specified)

GRANT

29

Granting the CONNECT privilege (GRANT statement with CONNECT specified)

30

Granting the schema definition privilege (GRANT statement with SCHEMA specified)

31

Granting the SELECT privilege (GRANT statement with SELECT specified)

32

Granting the INSERT privilege (GRANT statement with INSERT specified)

33

Granting the UPDATE privilege (GRANT statement with UPDATE specified)

34

Granting the DELETE privilege (GRANT statement with DELETE specified)

35

Granting the TRUNCATE privilege (GRANT statement with TRUNCATE specified)

36

Granting the REFERENCES privilege (GRANT statement with REFERENCES specified)

37

Granting the IMPORT TABLE privilege (GRANT statement with IMPORT TABLE specified)

38

Granting the REBUILD INDEX privilege (GRANT statement with REBUILD INDEX specified)

39

Granting the GET COSTINFO privilege (GRANT statement with GET COSTINFO specified)

40

Granting the EXPORT TABLE privilege (GRANT statement with EXPORT TABLE specified)

41

Granting the MERGE CHUNK privilege (GRANT statement with MERGE CHUNK specified)

42

Granting the CHANGE CHUNK COMMENT privilege (GRANT statement with CHANGE CHUNK COMMENT specified)

43

Granting the CHANGE CHUNK STATUS privilege (GRANT statement with CHANGE CHUNK STATUS specified)

44

Granting the ARCHIVE CHUNK privilege (GRANT statement with ARCHIVE CHUNK specified)

45

Granting the UNARCHIVE CHUNK privilege (GRANT statement with UNARCHIVE CHUNK specified)

46

Revoking the DBA privilege (REVOKE statement with DBA specified)

REVOKE

47

Revoking the CONNECT privilege (REVOKE statement with CONNECT specified)

48

Revoking the schema definition privilege (REVOKE statement with SCHEMA specified)

49

Revoking the SELECT privilege (REVOKE statement with SELECT specified)

50

Revoking the INSERT privilege (REVOKE statement with INSERT specified)

51

Revoking the UPDATE privilege (REVOKE statement with UPDATE specified)

52

Revoking the DELETE privilege (REVOKE statement with DELETE specified)

53

Revoking the TRUNCATE privilege (REVOKE statement with TRUNCATE specified)

54

Revoking the REFERENCES privilege (REVOKE statement with REFERENCES specified)

55

Revoking the IMPORT TABLE privilege (REVOKE statement with IMPORT TABLE specified)

56

Revoking the REBUILD INDEX privilege (REVOKE statement with REBUILD INDEX specified)

57

Revoking the GET COSTINFO privilege (REVOKE statement with GET COSTINFO specified)

58

Revoking the EXPORT TABLE privilege (REVOKE statement with EXPORT TABLE specified)

59

Revoking the MERGE CHUNK privilege (REVOKE statement with MERGE CHUNK specified)

60

Revoking the CHANGE CHUNK COMMENT privilege (REVOKE statement with CHANGE CHUNK COMMENT specified)

61

Revoking the CHANGE CHUNK STATUS privilege (REVOKE statement with CHANGE CHUNK STATUS specified)

62

Revoking the ARCHIVE CHUNK privilege (REVOKE statement with ARCHIVE CHUNK specified)

63

Revoking the UNARCHIVE CHUNK privilege (REVOKE statement with UNARCHIVE CHUNK specified)

64

Creating an HADB user (CREATE USER statement)

CREATE USER

65

Deleting an HADB user (DROP USER statement)

DROP USER

66

Changing user information for an HADB user (ALTER USER statement)

ALTER USER

67

Definition SQL event

Defining an index (CREATE INDEX statement)

CREATE INDEX

68

Defining a schema (CREATE SCHEMA statement)

CREATE SCHEMA

69

Defining a base table (CREATE TABLE statement)

CREATE TABLE

70

Defining a viewed table (CREATE VIEW statement)

CREATE VIEW

71

Deleting an index (DROP INDEX statement)

DROP INDEX

72

Deleting a schema (DROP SCHEMA statement)

DROP SCHEMA

73

Deleting a base table (DROP TABLE statement)

DROP TABLE

74

Deleting a viewed table (DROP VIEW statement)

DROP VIEW

75

Changing a table definition (ALTER TABLE statement)

ALTER TABLE

76

Changing a viewed table definition (ALTER VIEW statement)

ALTER VIEW

77

Data manipulation SQL event

Retrieving data from tables (SELECT statement)

SELECT

78

Inserting rows into tables (INSERT statement)

INSERT

79

Updating table row data (UPDATE statement)

UPDATE

80

Deleting table row data (DELETE statement)

DELETE

81

Deleting all row data from a table (TRUNCATE TABLE statement)

TRUNCATE TABLE

82

Deleting all row data in a chunk (PURGE CHUNK statement)

PURGE CHUNK

83

SQL parsing error

UNKNOWN

84

Acquiring data stored in chunks (#GETDATA subcommand of adbsql command)

GETDATA

85

Acquiring the number of data items stored in a chunk (#GETCOUNT subcommand of adbsql command)

GETCOUNT

86

Displaying table information (#TABLES subcommand of adbsql command)

TABLES

87

Displaying column information (#COLUMNS subcommand of adbsql command)

COLUMNS

88

Displaying index information (#INDEXES subcommand of adbsql command)

INDEXES

89

Displaying chunk information (#CHUNKS subcommand of adbsql command)

CHUNKS

90

Displaying authorization identifiers (#GETUSER subcommand of adbsql command)

GETUSER

91

Command operation event

Importing data (adbimport command)

ADBIMPORT

92

Rebuilding indexes (adbidxrebuild command)

ADBIDXREBUILD

93

Collecting cost information (adbgetcst command)

ADBGETCST

94

Analyzing DB status (adbdbstatus command)

ADBDBSTATUS

95

Exporting data (adbexport command)

ADBEXPORT

96

Merging chunks (adbmergechunk command)

ADBMERGECHUNK

97

Setting, changing, and deleting chunk comments (adbchgchunkcomment command)

ADBCHGCHUNKCOMMENT

98

Changing the chunk status (adbchgchunkstatus command)

ADBCHGCHUNKSTATUS

99

Archiving chunks (adbarchivechunk command)

ADBARCHIVECHUNK

100

Unarchiving chunks (adbunarchivechunk command)

ADBUNARCHIVECHUNK

101

Reorganizing system tables (base tables) (adbreorgsystemdata command)

ADBREORGSYSTEMDATA

102

Registering and deleting synonym dictionaries (adbsyndict command)

ADBSYNDICT

The following table shows the privilege names output in the PRIVILEGE_TYPE column in Table 12‒10: Column structure of table function derived table when retrieving audit trails.

Table 12‒12: Privilege names output in audit trails

No.

Privilege type

Privilege name output in audit trails

1

DBA privilege

DBA

2

CONNECT privilege

CONNECT

3

Schema definition privilege

SCHEMA

4

Audit admin privilege

AUDIT ADMIN

5

Audit viewer privilege

AUDIT VIEWER

6

SELECT privilege

SELECT

7

INSERT privilege

INSERT

8

UPDATE privilege

UPDATE

9

DELETE privilege

DELETE

10

TRUNCATE privilege

TRUNCATE

11

REFERENCES privilege

REFERENCES

12

IMPORT TABLE privilege

IMPORT TABLE

13

REBUILD INDEX privilege

REBUILD INDEX

14

GET COSTINFO privilege

GET COSTINFO

15

EXPORT TABLE privilege

EXPORT TABLE

16

MERGE CHUNK privilege

MERGE CHUNK

17

CHANGE CHUNK COMMENT privilege

CHANGE CHUNK COMMENT

18

CHANGE CHUNK STATUS privilege

CHANGE CHUNK STATUS

19

ARCHIVE CHUNK privilege

ARCHIVE CHUNK

20

UNARCHIVE CHUNK privilege

UNARCHIVE CHUNK

The following table shows the data types of data bound to dynamic parameters output in the PARAM column in Table 12‒10: Column structure of table function derived table when retrieving audit trails.

Table 12‒13: Output format for dynamic parameters

No.

Data type

Output format

1

INTEGER

Decimal

2

DECIMAL(m, n)

Decimal

3

SMALLINT

Decimal

4

DOUBLE PRECISION

Mantissa (decimal) and exponent (decimal)

5

CHARACTER(n)

'character-string'

6

VARCHAR(n)

'character-string'#

If the actual length is 0, the value is output as ''.

7

DATE

Predefined output representation for date data

8

TIME(p)

Predefined output representation for time data

9

TIMESTAMP(p)

Predefined output representation for time stamp data

10

BINARY(n)

X'hexadecimal-value'

11

VARBINARY(n)

X'hexadecimal-value'

If the actual length is 0, the value is output as X".

12

ROW

X'hexadecimal-value'
Legend:

m, n, and p: Positive integer

Note:

If the data bound to the dynamic parameter is a null value, NULL is output in the PARAM column.

#

If only a dynamic parameter is specified in the value expression of a NULL predicate and data that is not the null value is specified in the dynamic parameter, '*' is output regardless of the value that is specified.

The following table shows the output format of server definition entries in the SERVER_OPERAND column in Table 12‒10: Column structure of table function derived table when retrieving audit trails.

Table 12‒14: Output format of server definition

No.

Format

Output format

1

set format

operand-name = specified-value#1

2

Command format

command-name option-name specified-value[, option-name specified-value ...]#2
#1

The maximum length of the = specified-value portion is 1,024 bytes. Anything longer is truncated from the 1,025th byte.

#2

The maximum length of the option-name specified-value[, option-name specified-value ...] portion is 1,024 bytes. Anything longer is truncated from the 1,025th byte.

The following shows a specification example of the server definition, and an example of the information output in the SERVER_OPERAND column. A slash (/) delimits each valid pair of a server definition operand and its value in the output information.

■ Server definition specification example
set adb_db_path = XXXXX
set adb_rpc_port = YYYYY
set adb_sys_max_users = 10
set adb_sys_rthd_num = 40
set adb_sys_uthd_num = 128
set adb_sql_exe_max_rthd_num = 4
set adb_sys_memory_limit = 64000
adbbuff -g TBLBUF01 -n ADBUTBL01 -p 1000000 -v 1024
adbbuff -g IDXBUF01 \
        -n ADBUIDX01 \
        -p 2500000 
adbcltgrp -g group -m 10 -u 0 -r 40 -e 0 -w 0
■ Output example of SERVER_OPERAND column
adb_db_path = XXXXX / adb_rpc_port = YYYYY / adb_sys_max_users = 10 / adb_sys_rthd_num = 40 / adb_sys_uthd_num = 128 / adb_sql_exe_max_rthd_num = 4 / 
adb_sys_memory_limit = 64000 / adbbuff -g TBLBUF01 -n ADBUTBL01 -p 1000000 -v 1024 / adbbuff -g IDXBUF01 -n ADBUIDX01 -p 2500000 / adbcltgrp -g group -m 10 -u 0 -r 40 -e 0 -w 0

The following explains the values output for each event in the BEFORE_SYSTEM_INFO and AFTER_SYSTEM_INFO columns in Table 12‒10: Column structure of table function derived table when retrieving audit trails,

■ Values output when an event (adbstart or adbchgsrvmode) is successful

When an event (adbstart or adbchgsrvmode) is successful, the values shown in the following table are output to represent the HADB server operation mode:

Table 12‒15: HADB server operation mode values

No.

HADB server operation mode

Value

1

Normal mode

NORMAL

2

Quiescence mode

QUIESCENCE

3

Offline mode

OFFLINE

4

Maintenance mode

MAINTENANCE

When using the multi-node function, the BEFORE_SYSTEM_INFO column for the master node displays the HADB server operation mode for the node that was the master node the last time the HADB server started. The BEFORE_SYSTEM_INFO column for the slave node always displays NULL.

■ Values output when an event (adbchgnodetype) is successful

When an event (adbchgnodetype) is successful, the values shown in the following table are output to represent the node type:

Table 12‒16: Node type values

No.

Node type

Value

1

Master node

MASTER

2

Slave node

SLAVE
■ Values output when an event (adbchgsqltrc) is successful

When an event (adbchgsqltrc) is successful, the values shown in the following two tables are output to represent the trace information and trace level:

Table 12‒17: Trace information values

No.

Trace information

Value

1

Access path information

accesspath

2

Dynamic parameter information

param
Table 12‒18: Trace level value

No.

Trace level

Value

1

Output at SQL statement level

sql

2

Output at call level

call

When an event (adbchgsqltrc) is successful, the trace information and trace level that are in effect are output in that order, separated by a slash (/). If SQL tracing is disabled, a blank character is output.

An example of output when access path is in effect and the trace level is the SQL statement level is as follows:

accesspath / sql
■ Values output when an event (adbclientdefmang) is successful

When an event (adbclientdefmang) is successful, the client-managing definitions are output separated by slashes (/) in the command format explained in Table 12‒14: Output format of server definition. If centralized management of client definitions is disabled, a blank character is output.

The following are a specification example and output example of client-managing definitions.

  • Specification example of client-managing definitions

    adbclientmang -f client01.def -i USER01, USER02
adbclientmang -f client02.def -i USER03

  • Output example

    adbclientmang -f client01.def -i USER01, USER02 / adbclientmang -f client02.def -i USER03
■ Values output when an event (adbmodarea) is successful

When an event (adbmodarea) is successful, the values are output in the following format:

AREA NAME = DB-area-name, FILE NUMBER = number-of-DB-area-files

Description

DB-area-name

The DB area name. Either the specified value or a normalized character string is output.

number-of-DB-area-files

The number of DB area files. If the number of files is unknown, ? is output. If the DB area does not exist, 0 is output.

The following are a specification example and output example for a DB area addition and modification option file.

  • Specification example of DB area addition and modification option file

    adbaddarea -n ADBUTBL02

  • Output example (BEFORE_SYSTEM_INFO column)

    AREA NAME = ADBUTBL02, FILE NUMBER = 0

  • Output example (AFTER_SYSTEM_INFO)

    AREA NAME = ADBUTBL02, FILE NUMBER = 1
■ Values output when an event (adbmodbuff) is successful

When an event (adbmodbuff) is successful, the number of pages in the buffer for the local work table (5 to 100,000,000) changed by the adbmodbuff command is output.

The following are a specification example and output example for a buffer-modifying option file of the adbmodbuff command.

  • Specification example of buffer-modifying option file

    set adb_dbbuff_wrktbl_clt_blk_num = 256

  • Output example

    256
■ Values output when an event (adbcolumnize) is successful

When an event (adbcolumnize) is successful, the information about the status of the updated-row columnizing facility is output.

  • If the updated-row columnizing facility is enabled, ACTIVE is output.

  • If the updated-row columnizing facility is disabled, INACTIVE is output.

The following table shows the circumstances under which each object type is output in the OBJECT_TYPE column in Table 12‒10: Column structure of table function derived table when retrieving audit trails:

Table 12‒19: Circumstances in which object types are output in audit trails

No.

Object type

Output trigger

1

INDEX#1

  • Defining and deleting indexes

  • Rebuilding indexes (adbidxrebuild command)

2

SCHEMA

Defining and deleting schema

3

TABLE

  • Defining a base table

  • Changing the definition of a base table

  • Referencing a base table (including data export)

  • Updating a base table (including data import)

  • Deleting a base table

  • Granting or revoking an access privilege

  • Collecting cost information (adbgetcst command)

  • Merging chunks (adbmergechunk command)

  • Setting, changing, and deleting chunk comments (adbchgchunkcomment command)

  • Change the chunk status (adbchgchunkstatus command)

  • Archiving chunks (adbarchivechunk command)

  • Unarchiving chunks (adbunarchivechunk command)

  • Reorganizing system tables (base tables) (adbreorgsystemdata command)

4

TABLE FUNCTION#2

Referencing a table function derived table (including data export)

5

VIEW

  • Defining a viewed table

  • Re-creating a viewed table

  • Referencing a viewed table (including data export)

  • Updating a viewed table

  • Deleting a viewed table

  • Granting or revoking an access privilege
#1

Information about the indexes used to retrieve data from a table is not output.

#2

The object identifier is the function name of the table function. Arguments of the table function are not output.

The following table shows the number of rows output by each operation in the ACCESS_COUNT column in Table 12‒10: Column structure of table function derived table when retrieving audit trails.

Table 12‒20: Number of rows output in ACCESS_COUNT column

No.

Event

EVENT_SUBTYPE of audit trail output by operation

Number of output rows

1

SELECT statement

SELECT

Number of rows in search results#1, #2, #3

2

INSERT statement

INSERT

Number of rows#3 inserted into the table output in the OBJECT_NAME column in Table 12‒10: Column structure of table function derived table when retrieving audit trails

3

SELECT

NULL#2

4

UPDATE statement

UPDATE

Number of rows#3 updated in the table output in the OBJECT_NAME column in Table 12‒10: Column structure of table function derived table when retrieving audit trails

5

SELECT

NULL#2

6

DELETE statement

DELETE

Number of rows#3 deleted from the table output in the OBJECT_NAME column in Table 12‒10: Column structure of table function derived table when retrieving audit trails

7

SELECT

NULL#2

8

TRUNCATE TABLE statement

TRUNCATE TABLE

0

9

PURGE CHUNK statement

PURGE CHUNK

0

10

SELECT

NULL#2

11

Definition SQL statements

Subtype of each event

0#4

12

adbimport command

ADBIMPORT

Number of data rows#3 imported into the table output in the OBJECT_NAME column in Table 12‒10: Column structure of table function derived table when retrieving audit trails

13

adbexport command

ADBEXPORT

Number of exported data rows#2, #3

14

All other

Subtype of each event

NULL
#1

The output value is the number of rows for which FETCH processing had completed on the HADB server side when the cursor was closed. Due to batch transfer of search results, the output row count might be a number determined by the following formula instead of the number of rows actually undergoing FETCH processing on the client. In this case, the value will be larger than the number of rows undergoing FETCH processing on the HADB client side.

number-of-batch-transferred-rows × 2 - 1
#2

When executing one SQL statement or one command, if multiple audit trails are output for which the content of the EVENT_SUBTYPE column in Table 12‒10: Column structure of table function derived table when retrieving audit trails is the same, the same value will be output in the ACCESS_COUNT column for these audit trails.

#3

If the value of the EVENT_RESULT column in Table 12‒10: Column structure of table function derived table when retrieving audit trails is FAILURE, 0 is output.

#4

If multiple audit trails are output when one definition SQL statement is executed, the same value (0) is output in the ACCESS_COUNT column for these audit trails.

For details about the circumstances in which an audit trail is output for each event and a description of the output items, see 12.9.3 Audit trail output triggers and output items.

To Page Top

All Rights Reserved. Copyright (C) 2012, 2020, Hitachi, Ltd.