DB2 Server for VSE. Operations Guide / Handbooks (2004-2007)

Value-or-string is of the form:

[-]n...n if displaying decimal data (binary internally)

‘c...c’ if displaying (unconverted) character data

‘h...h’X if displaying hex data (binary internally)

v L/G_varname=HEXADECIMAL DUMP: (line 1)

0000 hhhhhhhh (up to 12 hex words) (line 2)

–

0030 hhhhhhhh (up to 12 hex words) (line 3 if needed).

More lines are included as needed for up to 32 kilobytes of hex data. For lines

to n, the first 4 characters are the hex offset of displayed hex dump data (for

example, 0 for the first line, 30 for the second line). The rest of the line is hex

output displayed in groups of 8 hex characters (each a binary word) with 12

groups per line (less only if end of output data reached).

Trace output may also contain printable character strings that are trace point

dependent and clarify processing being performed or the data being displayed.

Isolating Problems

You can use either the ISQLMAP or the ISQLTRACE commands to isolate

problems within the ISQL facility.

ISQLMAP

The ISQL facility contains its own load module map, which is invoked by the

command:

ISQLMAP

ISQL Dumps in DB2 Server for VSE

Use of CICS dumps in problem diagnosis is presented in the CICS Problem

Determination Guide

ISQL In-Core Trace

The ISQL facility lets you trace activity within ISQL. Items that are traced include:

v Calls to and returns from ISQL modules

v SQL return codes

v ISQL messages.

The trace is always active. It stores a limited amount of information internally in a

trace table.

The initial trace table holds 50 entries. After 50 entries are made, the table wraps

and only the last 50 entries to the table are kept. To change the size of the internal

table and select how trace information is to be displayed, issue the ISQLTRACE

command.

►►

ISQLTrace

DUMP

►◄

DISPLAY

DUMP

creates an unformatted storage dump hardcopy of the trace table.

Chapter 5. Tracing the DB2 Server for VSE & VM System

177

In a VSE system, the ISQL facility issues a CICS dump to the CICS dump data

set. You should use your installation’s usual job for printing the dump data set.

In a VM system, the ISQL facility issues a CP DUMP command. The dump

uses the printer with the lowest virtual address. The application server issues a

CP CLOSE PRINT to close the print file. This also closes any other open print

files.

DISPLAY

displays the formatted trace table on the display. If more than 50 (or n) entries

have been made into the trace table, the entries wrap. Only the last 50 (or n)

entries in the table are displayed. The entries are displayed last in, first out, in

the reverse of the order they were put in the table. Table 2 shows the internal

storage layout of these entries in DB2 Server for VM.

n to change the size of the trace table. Replace n with the number of trace entries

that are to be contained in the trace table. n must be a number from 50 to 1000.

Table 2. Internal Storage of the ISQL Trace Table on VM

TRMOD

TRCMOD

TRR15

TRLBL

TRCODE

TRCHAR

CHAR(8)

FIXED(31)

FIXED(15)

CHAR(1)

CHAR(5)

calling- returning

called module

address or return

2-byte label

C-calls or

5-bytes char mod

module

code

number

R-return

dep info

For Example:

ARIISTR

ARIISQL1 (hex 0)

0048A808

0002

0001

C R

SEL EXITS

ARIISQL1

00000004

TRMOD

is the module issuing the call or return.

TRCMOD

is the module being called.

TRR15

holds the value of register 15 at the time of the entry, which is the address of

the called module or the return code.

TRLBL

is the number associated with the label of the CALL or RETURN:

A label of TRLC7 would have a TRLBL number = 7, and would be the 7th

CALL issued from the module.

A label of TRLR3 would be the third RETURN issued from the module, and

would have a TRLBL number = 3.

TRCODE

C indicates a CALL label, R indicates a RETURN label.

TRCHAR

contains module-dependent information. The information saved here varies for

each module. It is a maximum of 5 bytes.

The following is an example output of the ISQLTRACE DISPLAY command. The

addresses, label numbers, and character information are not shown in the example.

178

Operation

ARI7181I ISQL trace table display.

First entry: 2B36F0 Current entry: 2B3C50 Last entry: 2B3D10

ARIIPSQ CALLED ARIITKN,R15= 443E60,TRLBL= 0001,TRCHAR= TKN1

RETURN FROM ARIITKN,R15= 000000,TRLBL= 0003,TRCHAR=

ARIIPSQ CALLED ARIITKN,R15= 443E60,TRLBL= 0015,TRCHAR= TKN2

RETURN FROM ARIITKN,R15= 000000,TRLBL= 0003,TRCHAR=

ARIIPSQ CALLED ARIIPQY,R15= 44C8B0,TRLBL= 0018,TRCHAR= SEL

ARIIPQY CALLED ARIICNV,R15= 458860,TRLBL= 0001,TRCHAR= CONS

RETURN FROM ARIICNV,R15= 000010,TRLBL= 0001,TRCHAR= DECC

ARIIPQY CALLED ARIIGM ,R15= 44220E,TRLBL= 0031,TRCHAR= SQLDA

Figure 90. A Typical Example ISQLTRACE DISPLAY Output

Appendix A. DB2 Server for VSE & VM Initialization

Parameters

This section describes the parameters you can specify when starting the application

server. For a complete description of the initialization parameters, and information

on initialization parameter values, talk to your system administrator or see the DB2

Server for VSE System Administration, or DB2 Server for VM System Administration

manual.

Common Parameters

The following table contains initialization parameters common to both the VSE

environment and the VM environment. Initialization parameters specific to each

system follow.

Parameter

Default

Description

ALTLOG=N|Y

Alternate Logging

This 1-character field indicates whether an inactive log

exists in the system. Y indicates an inactive log is

present; N indicates no inactive log has been defined.

The inactive log will only be used when LOGMODE=L.

When ARCHPCT is reached, an attempt will be made to

switch to the inactive log rather than take a log archive.

This parameter is used only when STARTUP=L or

STARTUP=C is specified.

Note: The DB2 Server for VM EXECs always supply the

ALTLOG parameter for you. You should not specify the

ALTLOG parameter directly.

179

Parameter

Default

Description

ARCHPCT=n

Archive Percent

n is a decimal value indicating percentage. When the log

is n percent full, the database manager begins an archive

process, prompting the operator to mount the required

tape. The n can be any number from 10 to 99.

This parameter is used only when LOGMODE=A or

when LOGMODE=L is specified. When LOGMODE=A,

a database archive is started when the ARCHPCT value

is reached. When LOGMODE=L and ALTLOG=N, a log

archive is started instead of a database archive. When

LOGMODE=L and ALTLOG=Y, an attempt is made to

switch to an inactive log disk. For optimum use of the

archive function, the value of ARCHPCT should be less

than the value of SLOGCUSH.

CHARNAME=name

After

Character Set Name

installation:

INTERNATIONAL

This 1- to 18-character field specifies the name of the

character set that the database manager uses to define

After

the system language. The name you specify must

migration:

correspond to the name of a character set that has been

ENGLISH

defined in the NAME column of the

SYSTEM.SYSCHARSETS table and the CHARNAME

column of the SYSTEM.SYSCCSIDS table.

The value of the CHARNAME parameter specified in

the SQLSTART EXEC or startup jcl determines the

values of the CHARNAME, CCSIDSBCS, CCSIDMIXED

and CCSIDGRAPHIC rows in the

SYSTEM.SYSOPTIONS table. It also updates the CCSID

and SUBTYPE attributes for the character columns of

the system catalog tables. If CHARNAME is not

specified, the CHARNAME value is determined from

the CCSID value of the CNAME column in the

SYSTEM.SYSCOLUMNS table. For information on

selecting a CHARNAME and the related performance

considerations, see the DB2 Server for VSE System

Administration, or DB2 Server for VM System

Administration manual.

CHKINTVL=n

Checkpoint Interval

n is a decimal value that defines when the database

manager is to automatically take a checkpoint. A

checkpoint is taken after every n log pages have been

written. n can be any number from 1 to 99,999,999.

DBSPNUM=n

n indicates the dbspace number on which release empty

pages processing will be performed. n must be an

acquired dbspace number in the database or *.

DBSPNUM is only valid when SYSMODE=S and

STARTUP=P.

Note: The DB2 Server for VM EXECs always supply the

DBSPNUM parameter for you. You should not specify

the DBSPNUM parameter directly.

180

Operation

Parameter

Default

Description

DISPBIAS=n

Dispatcher Bias

n is a decimal value from 1 to 10. The DISPBIAS

parameter controls how the DB2 Server for VSE & VM

dispatcher operates. When the DISPBIAS parameter is

set closer to 10, the dispatcher operates more like a

priority dispatcher. When the DISPBIAS parameter is set

closer to 1, the dispatcher operates more like a round

robin dispatcher. Refer to the DB2 Server for VSE System

Administration, or DB2 Server for VM System

Administration manual for more information on the

DISPBIAS parameter.

DSPSTATS=nn

This 2 character field allows values 0, 1 and 2 for option

1 and values 0 and 1 for option 2. If the first option is 1,

checkpoint performance data is displayed on the console

in the form of messages every time a checkpoint occurs.

If the first option is 2, checkpoint performance detailed

data is displayed on the console in the form of messages

every time a checkpoint occurs. If the second option is

1, the COUNTER * command will be internally invoked

when the database manager is shutdown. This displays

the counter value for the database manager at the last

possible moment. If VMDSS is activated, the COUNTER

POOL * command is also issued.

DUALLOG=N|Y

Dual Logging

This 1-character field indicates whether the database

manager is to maintain an identical copy of the active

log and, if ALTLOG=Y, a backup copy of the alternate

log. Y indicates dual logging; N indicates single logging.

This parameter is used only when STARTUP=L or

STARTUP=C is specified.

Note: The DB2 Server for VM EXECs always supply the

DUALLOG parameter for you. You should not specify

the DUALLOG parameter directly.

DUMPTYPE=P|F|N

Dump Type

This 1-character field indicates the type of dump to be

taken if the application server ends abnormally. F

indicates a full dump. For VM, this is the entire virtual

machine. For VSE, this is the entire partition. P indicates

a partial dump, where everything is dumped except

DB2 Server for VSE & VM code. (DB2 Server for VSE &

VM control blocks are included in the dump.) N

indicates no dump. No dump is taken if a DB2 Server

for VSE & VM limit error or a hardware error occurs or

if the termination is because of a user specification error.

ATTENTION: IBM recommends that you always use

DUMPTYPE=F, as most problems cannot be resolved

with a partial dump. This will help eliminate the need

for recreating problems just to get a full dump.

Note: In single user mode, DUMPTYPE=P includes the

application code in the dump.

Appendix A. DB2 Server for VSE & VM Initialization Parameters

181

Parameter

Default

Description

EXTEND=Y|N

Filtered Log Recovery

This 1-character field indicates to the database manager

whether it should read EXTEND input file commands. Y

means that the database manager should read EXTEND

input file commands, and N, the default, means it

should not. You can specify Y only when STARTUP=W,

R, U, or F (F is valid for DB2 Server for VSE only).

(Otherwise, you receive an error message.)

Specify the EXTEND input file commands in either a

CMS file that has fixed-length, 80-byte records, or in

SYSIPT. Before starting the application server enter a

CMS FILEDEF command with the ddname ARIEXTND

that identifies the CMS file. The EXTEND input file

commands are described in the DB2 Server for VSE &

VM Diagnosis Guide and Reference manual.

If you are starting the application server in single user

mode, the single user mode job must not use SYSIPT.

182

Operation

Parameter

Default

Description

Logging Mode

LOGMODE=

This 1-character field indicates how the database

A|L|N|Y

manager is to maintain the logs:

The database manager is to maintain the logs

and automatically archive the database when

the level specified by ARCHPCT is reached.

The database manager is to maintain the logs

and automatically archive the log or switch to

the inactive log disk (if alternate logging is

enabled)when the level specified by ARCHPCT

is reached.

The database manager is not to maintain the

logs for recovery. A log is still needed, however,

for internal application server operations. A log

is always needed, whether for recovery of user

data or not. The physical device used for the

log must always be allocated, even when

LOGMODE=N is specified. You cannot specify

LOGMODE=N when SYSMODE=M.

The database manager is to maintain the logs.

It is not always possible to switch log modes. Refer to

the DB2 Server for VM System Administration, or DB2

Server for VSE System Administration manual for

information about switching log modes.

For VSE:

Note that if A or L is specified, DB2 Server for VSE job

control must include a TLBL statement with the file

name ARIARCH for the database archive tape. If you

specify LOGMODE=L, the job control must also include

a TLBL statement with the file name ARILARC for the

active log archive tape. If alternate logging is enabled, a

TLBL statement with the file name ARILALT will be

needed for the inactive log archive tape. The database

manager will dynamically assign the tape units unless

you specified static assign in your startup JCL. See

Figure 21 on page 47 for sample JCL. When dynamic

assign is chosen, your JCL must not contain an ASSGN

statement.

For VM:

The SQLSTART EXEC issues a CMS FILEDEF command

with the ddname ARIARCH for the database archive file

(which can be directed only to tape). It also issues a

CMS FILEDEF command with the ddname ARILARC

for the active log archive file (which can be directed to

tape or disk). If alternate logging is enabled, SQLSTART

also issues a CMS FILEDEF command for the ddname

ARILALT for the inactive log archive file (which can

also be on tape or disk). Users can enter CMS FILEDEF

and LABELDEF commands for the archive file before

starting the database manager. (User FILEDEFs override

the FILEDEFs that SQLSTART issues.) The ddnames

must be ARIARCH for database archive files and

ARILARC and ARILALT for log archive files.

Appendix A. DB2 Server for VSE & VM Initialization Parameters

183

Parameter

Default

Description

LTIMEOUT=n

Lock Timeout

Number of seconds before a lock wait will timeout.

This initialization parameter specifies a general lock

wait timeout period for any SQL application, and

especially as the way to avoid global deadlocks for

DUOW applications.

The range of the LTIMEOUT value is 0 to 99999

seconds.

The default is 0, which means no timeout.

NCSCANS=n

Number of Concurrent Scans

n is a decimal value that determines the maximum

number of concurrent scans that a given user might

have outstanding. A scan maintains positioning

information within the database. That is, to hold

information related to a DB2 Server for VSE & VM

cursor. The information is kept in a scan table. The

maximum size of this scan table is 32 kilobytes. The

maximum number of DB2 Server for VSE & VM cursors

per active agent is 255. This value might be less

depending on the amount of positioning information

required for a particular cursor. The average scan for a

cursor is 50 bytes. n can be any number from 1 to 655.

For more information, see the DB2 Server for VSE System

Administration, or DB2 Server for VM System

Administration manual.

NCUSERS=n

Number of Concurrent Users

n is a decimal value that defines the maximum number

of concurrent users that the database manager can

support in multiple user mode. The actual number of

logged-on users, however, might exceed this number.

One of the uses of this value is to generate DB2 Server

for VSE & VM task control blocks called agent structures

(or agents). Normally, one user is associated with one

agent structure.

NDIRBUF=n

NPAGBUF

Number of Directory Buffers

n is a decimal value that defines how many directory

buffers are to be reserved in virtual storage. These

buffers are shared by all users. Each buffer requires 512

bytes. n can be any number from 10 to 400000.

184

Operation

Parameter

Default

Description

NLRBS=n

Calculated

Number of Lock Request Blocks for the Application

Server

n is a decimal value that defines the maximum number

of lock request blocks available in multiple user mode.

This value limits the total number of locks that can be

requested by all concurrent users. When the application

server approaches this value (comes to within

2*NCUSERS of the value), the lock escalation process

occurs. The default value is NCUSERS*2 +

(NCUSERS*NLRBU)/2 + 10. The minimum value is the

greater of 50 or NCUSERS*2. The maximum value is 583

333.

NLRBU=n

1000

Number of Lock Request Blocks per User

n is a decimal value that defines the maximum number

of lock request blocks available to any one user in

multiple user mode. When the application server

reaches this value for any application, the lock

escalation process occurs. (See the LOCKLMT and

ESCALATE counters under the COUNTER command on

page 58 for more details about lock escalation.) In

general, you can assume that for every lock that an

application holds, two lock request blocks are required.

The minimum value is 10, and the maximum is 583,333.

NPACKAGE=n

Maximum Number of Active Packages in a LUW

n is a decimal value that determines the maximum

number of active packages in an LUW. The value of n

must be in the range from 1 to m where (m x NCUSERS)

must be less than 32767. For more information see the

DB2 Server for VSE System Administration, or DB2 Server

for VM System Administration manual.

NPACKPCT=n

Packages in Cache Threshold

n is an integer value indicating the number of packages

which will remain loaded in cache as a percentage of

the total cache. n can be in the range of 0 to 100 percent.

Size of package cache=

NPACKAGE x NCUSERS

Threshold =

size of package cache

x NPACKPCT / 100

When the number of loaded packages exceeds the

threshold, packages will be discarded at the end of the

LUW until the number of loaded packages drops below

the threshold level. The value of the threshold is

truncated to an integer. If the threshold is 0, one

package will be kept in cache. For more information see

the DB2 Server for VSE System Administration, or DB2

Server for VM System Administration manual.

Appendix A. DB2 Server for VSE & VM Initialization Parameters

185

Parameter

Default

Description

NPAGBUF=n

Calculated

Number of Data Page Buffers

n is a decimal value that defines how many data page

buffers are to be reserved in virtual storage. These

buffers are shared by all users. Each buffer requires 4

096 bytes. The default value for n is (10 + (4 x

NCUSERS)). n can be any number from 10 to 400,000.

PARMID=name

None

Name of Parameter File

A 1- to 8-character name that defines a DB2 Server for

VSE source member or a DB2 Server for VM CMS file

that contains initialization parameters that are to

override the default values. (The VM file type must be

SQLPARM.)

PROCMXAB=nnn

Number of Stored Procedures

nnn is a decimal value specifying the number of times a

stored procedure is allowed to terminate abnormally,

after which a STOP PROC ACTION REJECT is

performed against the procedure and all subsequent

SQL CALL statements are rejected. Note that a timeout

that occurs while waiting for a stored procedure server

to be assigned for an SQL CALL statement is not

included in this count. The default, 0, means that the

first abend of a stored procedure causes SQL CALLs to

that procedure to be rejected. For production systems,

you should accept the default.

PROGNAME=name

None

Program Name

A 1- to 8-character name that defines the program to be

loaded and called by the application server when the

application server is initialized. The name is required for

single user mode only when STARTUP=W, U, F, or R. (F

is valid for DB2 Server for VSE only.) It can be the name

of an application, the DBS Utility program, or a DB2

Server for VSE & VM preprocessor.

PTIMEOUT=nnnnn

180

PSERVER Timeout

This parameter serves two purposes:

1. The number of seconds before DB2 Server for VSE &

VM ceases to wait for an SQL CALL to be assigned

to a stored procedure server. If the PTIMEOUT

interval expires, the SQL statement fails, and

SQLCODE -913 is returned with SQLSTATE 40001.

2. The number of seconds before DB2 Server for VSE &

VM ceases to wait for the START PSERVER

command to complete. If the PTIMEOUT interval

expires, message ARI4168I is displayed and the

START PSERVER command terminates.

A value of 0 means that no PTIMEOUT is in effect.

186

Operation

Parameter

Default

Description

SLOGCUSH=n

DB2 Server for VSE & VM Log Cushion

n is a decimal value indicating percentage. When the log

becomes n percent full, the database manager invokes

the log overflow procedure. The log overflow procedure

is described in the DB2 Server for VSE System

Administration, and DB2 Server for VM System

Administration manuals. n can be any number from 11 to

100. For optimum use of the archive function, the value

for SLOGCUSH should be greater than the value of the

ARCHPCT parameter.

SOSLEVEL=n

Short-on-Storage Level

n is a decimal value indicating percentage. For

LOGMODE=N, message ARI0202I is issued when n

percent of the storage pool pages remain free. For every

other LOGMODE, when n percent or less of the pages

in the storage pool are free, the database manager takes

a checkpoint in an attempt to free shadow pages. If,

after this checkpoint, there are still only n percent or less

free pages in the storage pool, message ARI0202I is

issued once. When this message occurs, the system

should be shut down as soon as possible so that more

disk storage can be added to the particular storage pool

(with the ADD DBEXTENT function). n can be any

number from 1 to 100.

Appendix A. DB2 Server for VSE & VM Initialization Parameters

187

Parameter

Default

Description

STARTUP=type

Type of Startup

This 1-character field indicates how the application

server is to be started:

Database or catalog generation (cold start). The

database manager formats the entire database,

including the log.

Add dbextents, delete dbextents, or both.

Restore the database from an archive tape

without reformatting the database files. This

can be specified only when none of the

database files (BDISK, DDSK1-n) have been

replaced. Type F is valid in VSE only.

Reorganize the indexes on the catalog tables.

Redefine the log data sets (COLDLOG). The

LOGMODE must be set to Y.

Migrate a DB2 Server for VSE & VM database

from a previous to a current version of the

product.

Release empty pages.

Restore the database from an archive tape.

Add new dbspaces.

User restore. Use this parameter to start the

database manager if you have just restored the

database from a user archive.

The storage pool will be brought up-to-date

with the rest of the database if you are using

the Data Restore Feature. See the DB2 Server for

VSE & VM Data Restore for more information.

Normal start (warm start).

Cold start (C) or COLDLOG (L) cancels the restart

recovery procedure.

Warm start (W), archive restore (R or F), or user restore

(U) can be specified in either multiple user or single

user mode. The other options can be specified only in

single user mode (SYSMODE=S).

Note: In VM, you should not normally specify

STARTUP=C, L, E, I, M, or S. The database manager

provides EXECs for each of these functions. The startup

parameters are used as part of a process to perform each

of the functions, but the parameters do not, in

themselves, perform the entire function.

For descriptions of the EXECs for values C, E, L, R, S, U

and W , see the DB2 Server for VM System Administration

manual; for values I and P see the DB2 Server for VSE &

VM Database Administration manual, and for value M see

the DB2 Server for VM Program Directory.

188

Operation

Parameter

Default

Description

SYNCPNT=Y|N

Sync point manager (SPM) support

This 1-character field indicates whether or not a sync

point manager (SPM) will be used to coordinate

DRDA-2 DUOW two-phase commit and

resynchronization activity.

Y means that the server will use a sync point manager,

if possible, to coordinate the two-phase commits and

resynchronization activity.

N means that the server will not use an SPM to perform

two-phase commits. Also if N is specified, DB2 Server

for VSE & VM is limited to multi-read, single-write

distributed units of work and it can be the single-write

site.

Note: If Y is specified, but DB2 Server for VSE & VM

finds that a sync point manager is not available, then the

server will operate as if N was specified.

SYSMODE=M|S

System Mode

This 1-character field indicates whether the application

server is to run in single user mode (S) or multiple user

mode (M).

SECALVER=Y|N

Security

The value of this parameter is determined by the value

of the SECTYPE parameter. If SECALVER=Y is specified,

and the requester sent a userid but no password for

authentication, the application server assumes that the

user is already verified. If SECALVER=N is specified,

the application server assumes that the user is not

already verified and a password or PassTicket must

accompany a userid for authentication.

This parameter is valid only when the incoming connect

request is via TCP/IP from any type of requester or

when the incoming connect request is from a DB2

Server for VSE DRDA application requester via SNA.

SECTYPE=DB2|ESM

DB2

Security

If SECTYPE=DB2 is specified, the

SYSTEM.SYSUSERAUTH table is checked if the user has

connect authority; the password column is not checked.

If SECTYPE=ESM is specified, a check is performed

during initialization to ensure that the external security

manager is available and that it supports the

RACROUTE interface.

This parameter is valid only when the incoming connect

request is via TCP/IP from any type of requester or

when the incoming connect request is from a DB2

Server for VSE DRDA application requester via SNA.

Appendix A. DB2 Server for VSE & VM Initialization Parameters

189

Parameter

Default

Description

TAPEMGR=N|Y

Tape Manager

This 1 character field indicates whether there is a tape

manager available to handle tape assigns during

database and log archives. Y indicates there is a tape

manager; N indicates there is no tape manager.

If TAPEMGR=N, then the operator will be prompted to

enter the virtual device address of the archive output.

If TAPEMGR=Y, then the tape assign will automatically

be handled by the tape manager and the operator will

not be prompted to enter the virtual device number

(cuu). In VM, log archives will automatically be directed

to tape if TAPEMGR = Y.

TCPPORT=nnnnn

This identifies the TCP/IP port number as an integer in

the range of 1 to 65535.

If TCPPORT=0 is specified, no TCP/IP initialization will

be attempted.

For VSE, if TCPPORT is not specified, the DBNAME

directory will determine whether TCP/IP initialization is

required and what port number should be used. For

VM, if TCPPORT is not specified, the ETC SERVICES

file is used to determine what port number should be

used.

Note: For VSE, the same TCP/IP port number can be

used among different application servers to initialize

TCP/IP support. In this case, the last server using the

duplicate port number will become the target server

owning the port. All later incoming connect requests

using this port number will be connected to this target

server. To avoid misdirected connection requests,

duplicate port numbers should not be used.

TCPMAXRT=n

158

Maximum number of TCP/IP support retry attempts.

n is the maximum number of times that automatic

restart of TCP/IP support is attempted.

The range of the TCPMAXRT value is 1 to 9999.

The default is 158.

Note: If the TCPRETRY parameter value is N, the value

of TCPMAXRT is ignored by the database manager.

190

Operation

Parameter

Default

Description

TCPRETRY=Y|N

Automatic restart if TCP/IP support fails.

This 1 character field indicates whether or not the

database manager will try to automatically restart

TCP/IP support if it fails.

Y indicates the database manager will try to

automatically restart TCP/IP support if it fails.

N indicates the database manager will not try to

automatically restart TCP/IP support if it fails.

N indicates the database manager will not try to

automatically restart TCP/IP support if it fails.

Note: If Y is specified, but TCPPORT = 0, then the

database manager will change the value to N.

Common Tracing Parameters

During DB2 Server for VSE initialization, the application server prompts you for

the tape cuu or DISK. The DISK reply causes the application server to open a

DASD SAM file for tracing. If any tracing is to be done, DB2 Server for VSE job

control must include a TLBL or DLBL statement with the file name ARITRAC.

During DB2 Server for VM initialization, the SQLSTART EXEC supplies this

(default) CMS FILEDEF command: FILEDEF ARITRAC TAP2 SL (BLKSIZE 4096

NOCHANGE PERM

Parameter

Default

Description

TRACDBSS= nnnnnnnnnnn

Zeros

Tracing in DBSS

This 11-digit parameter specifies the parts of the DBSS

that are to be traced and the level. Where 0 is specified,

tracing is not done. Where 1 is specified, tracing

displays only module entry and exit points. Where 2 is

specified, tracing is done in greater detail. The positional

digits correspond to the following DBSS functions and

subcomponents: ENTRY, EXIT, LOG, LOCK, LUW, DC,

DM, STOR, SORT, INDEX, and STAT. (A description of

these is in “Starting a Trace” on page 160.)

TRACDRRM=nnnn

Zeros

Tracing of the DRRM component

This four-digit parameter specifies the parts of the

DRRM that are to be traced and the level. Where 0 is

specified, tracing is not done. Where 1 is specified,

tracing displays only module entry and exit points.

Where 2 is specified, tracing is done in greater detail.

The positional digits correspond to the following DRRM

functions and subcomponents: the first is the DDM

Parser, the second is the DDM GENERATOR, the third

is the DDM Data Dictionary, and the fourth is the

RDIIN Manager. (A description of these is in “Starting a

Trace” on page 160.)

Note: You will only receive trace output for the work

unit manager if you specified AUTO for the

PROTOCOL parameter.

Appendix A. DB2 Server for VSE & VM Initialization Parameters

191

Parameter

Default

Description

TRACDSC=nn

Zeros

Tracing of the DSC component

This two-digit parameter specifies the parts of the DSC

component to be traced and the level. Where 0 is

specified, no tracing is done. Where 1 is specified,

tracing is done in limited detail. Where 2 is specified,

tracing is done in greater detail. The positional digits

correspond to the following DSC functions and

subcomponents: AGENT and COM. (A description of

these is in “Starting a Trace” on page 160.)

TRACEBUF=n

Zero

Tracing in Memory

The parameter n specifies the amount of memory in

kilobytes that you want to allocate to the trace buffer.

The range of the TRACEBUF integer value n is 0 to

99999. A suggested size for the trace buffer is 100

kilobytes or more. It is recommended that you increase

either the Virtual Machine memory or the size of the

partition by the size of the trace buffer before you start

a trace in memory. The trace buffer will only be created

if you specify TRACEBUF with at least one of the

startup initialization parameters TRACCONV,

TRACRDS, TRACDBSS, TRACDSC, TRACDRRM,

TRACSTG, or TRACWUM. Trace records are stored in

wrap-around mode in a fixed size buffer. A trace buffer

will not be created if the TRACEBUF default or n=0 is

selected.

If there are records in the buffer, they are written to the

trace output file by the TRACE OFF command or by

DB2 Server for VSE & VM shut down processing. For

details see SQLEND and TRACE OFF operator

commands on pages 11 and 165 respectively.

Note: If there is insufficient storage available for the

trace buffer, error message ARI0155E will be displayed

on the operator console and the system will not be

started.

TRACRDS=nnnnnnn

Zeros

Tracing in RDS

This seven-digit parameter specifies the parts of the RDS

to be traced and the level. Where 0 is specified, tracing

is not done. Where 1 is specified, tracing displays only

module entry and exit points. Where 2 is specified,

tracing is done in greater detail. The positional digits

correspond to the following RDS functions and

subcomponents: EXEC, PA, OPT, AG, INT, AU, and SG.

(A description of these is in “Starting a Trace” on page

160.)

TRACSTG=n

Zero

Tracing in the System and Working storage

This parameter can be specified on the application

server side only. It can be invoked during startup when

the database is first initialized and with the operator

TRACE command. The TRACSTG parameter has no

subcomponents. Values of n are 0 which specifies no

trace and 1 which specifies a level 1 trace.

192

Operation

Parameter

Default

Description

TRACWUM=n

Zero

Tracing in the Work Unit Manager

This parameter specifies the level of tracing for the work

unit manager. Where 0 is specified, tracing is not done.

Where 1 is specified, tracing is done for module entries

and exits. Where 2 is specified, tracing is done in greater

detail.

Note: You will only receive trace output for the work

unit manager if you specified AUTO for the

PROTOCOL parameter.

TRACCONV=n

Zero

Tracing of the CONV component

This parameter n specifies the level of tracing for the

CONV component. Where 0 is specified, tracing is not

done. Where 1 is specified, tracing is done in limited

detail. Where 2 is specified, tracing is done in greater

detail.

DB2 Server for VSE Parameters

The following table contains initialization parameters specific to the VSE

environment.

Parameter

Default

Description

ACCOUNT=T|D|E|N

Accounting

This 1-character field indicates whether the database

manager is to generate accounting records. N indicates

no accounting records are to be generated. D, E or T

indicates that you want the database manager to

generate accounting records. D means that you want the

database manager to write the records to a sequential

DASD file. E means that you want the database

manager to write records to a VSAM ESDS file. T means

that you want the database manager to write the records

to a tape file.

If you specify T, D, or E, your job control for starting the

application server must include appropriate TLBL,

DLBL, EXTENT, and ASSGN statements. The DB2 Server

for VSE System Administration manual describes how to

set up accounting files and appropriate job control.

ARCHTAPE=UNL|REW

REW

Archive Tape

This field controls when archive tapes are unloaded.

UNL indicates that at the end of writing to each tape of

a log or database archive, the tape will be unloaded

from the tape drive. This occurs regardless of whether

the archives are dynamically or statically assigned.

REW indicates that only archives taken using dynamic

allocation will be unloaded at end of tape. If it is the

last tape of the archive, and end of tape has not been

reached, the tape will remain in the drive and must be

unloaded manually by the operator.

Appendix A. DB2 Server for VSE & VM Initialization Parameters

193

Parameter

Default

Description

DBNAME=server-name

SQLDS

Server-Name

This optional 1- to 18-character field specifies the name

of the application server to be accessed. It is used by the

database manager to verify that the application

requesters are accessing the intended application server

when they attempt to connect to the database machine.

If the DBNAME is not specified, the partition or system

default will be assumed.

DBPSWD=password

None

Database Password

The password is a 1- to 8-character field that provides

the VSAM password for the database. If a VSAM

CONTROLPW or MASTERPW password was defined

for the DB2 Server for VSE data sets at database

generation, the password must match it exactly. This

value is not displayed with the others during

initialization.

DSPLYDEV=L|C|B

Display Device

This one-character field indicates where output such as

the initialization parameter values and the minidump is

to be routed. L indicates SYSLST. C indicates SYSLOG. B

indicates both SYSLST and SYSLOG.

RMTUSERS=nnnnn

Maximum number of remote users accessing the

application server

This parameter lets you specify the number of remote

users that can access the application server. The

allowable range of users is 0 to 65535; the practical

value of RMTUSERS depends on the availability of

virtual storage.

When a value has not been specified or nnnnn has the

value zero, remote users will not be able to access the

application server.

194

Operation

Parameter

Default

Description

TCPDISPB=n

TCP/IP Dispatcher Bias

This value specifies how frequently the TCPIP agent is

dispatched when there are no agents ready to run and a

system wait must be done. The TCPIP agent will be

dispatched every n times a system wait must be done.

Otherwise, the system wait is performed by the

DB2/VSE dispatcher.

Valid values are 1 to 100.

Setting the value to a high number can improve the

performance of agents that are doing I/O intensive

queries and are not using TCP/IP as the

communications protocol. This setting will not affect the

performance of CPU intensive queries as much as it will

affect I/O intensive queries.

Setting the value to a high number can degrade the

performance of users that are using TCP/IP as the

communications protocol.

This value is ignored if RMTUSERS=0. It is only active if

TCP/IP support is enabled.

DB2 Server for VM Parameters

Refer to “Starting the Application Server in a VM Environment” on page 2 for the

format of the SQLSTART EXEC syntax diagram and an explanation of the

SQLSTART EXEC options.

Parameter

Default

Description

ACCOUNT=D|N

Accounting

This 1-character field indicates whether the database

manager is to generate accounting records. N indicates

no accounting records are to be generated. D indicates

that you want the database manager to generate records.

The database manager writes (spools) the accounting

records to the VM system accounting file.

Appendix A. DB2 Server for VSE & VM Initialization Parameters

195

Parameter

Default

Description

DBMODE=L|G|N

Defined in VM directory

Database Mode

This 1-character field specifies the mode in which the

database manager is used. This parameter is only used

for VM/SP and VM/ESA systems.

If DBMODE=L is specified, the database manager

attempts to start the database as a local resource. There

must be a directory entry that allows that resource to be

identified as a local resource.

If DBMODE=G is specified, the database manager

attempts to start the database as a global resource. There

must be a directory entry that allows that resource to be

identified as a global resource.

DBMODE=N is used during migration; see the DB2

Server for VM System Administration manual for

information on DBMODE=N.

If the DBMODE parameter is not specified, the database

manager system defaults to the resource authorization

specified in its VM directory.

PROTOCOL=type

SQLDS

Protocol Type

Specifies the type of protocol that the application server

is capable of handling. Use SQLDS if your database is

only to be accessed through SQLDS protocol from DB2

Server for VM application requesters only. Specify the

AUTO option if both DRDA and SQLDS protocols are to

be supported.

Note: The DRDA protocol is not available if you have

not installed the optional DRDA code. For a description

of how to do this, refer to the DB2 Server for VM System

Administration manual.

TCPCAN=Y|N

Cancel TCP/IP Agent in VM Support

This 1-character field indicates whether he Cancel

TCP/IP Agent in VM Support is enabled or disabled. Y

indicates that the cancel support is enabled. N indicates

that the cancel support is disabled.

There is some performance overhead if cancel support is

enabled. This overhead may be unacceptable for some

users.

DB2 Server for VM Data Spaces Support Parameters

The following additional initialization parameters may be used when starting the

VM Server if you have installed the optional VMDSS code to support VM Data

Spaces.

Parameter

Default

Description

MAPPING=L|P

Sets whether the database manager will use logical or

physical mapping. To use logical mapping set

MAPPING=L. For physical mapping set MAPPING=P.

196

Operation

Parameter

Default

Description

SAVEINTV=n

Sets the save interval for the database to n blocks of 4KB

pages (32 pages in a block). The save interval is the

number of blocks in a data space that the database

manager modifies before it writes them to DASD. Valid

values are integers from 1 to 999999 blocks.

SEPINTDB=Y|N

Sets whether the database manager will use separate

unmapped internal dbspaces or not. To use unmapped

internal dbspaces, set SEPINTDB=Y. Note, setting

SEPINTDB=N does not turn Data Spaces Support off.

Because mapped internal dbspaces are assigned to one

storage pool, they will use whichever DASD I/O system

is specified for that pool.

TARGETWS=n

Sets the target working storage size for the database to n

megabytes. This limits the amount of main and

expanded storage that the database machine will use.

While the database machine may exceed this target, the

database manager will keep you at or below it if

possible. Valid values are integers from 1 to 999999

megabytes.

Appendix A. DB2 Server for VSE & VM Initialization Parameters

197

Appendix B. VMDSS Operator Commands

This section contains descriptions of the operator commands included with

VMDSS. These commands can be used in a VM/ESA only environment.

Note: You can also use the SHOW and COUNTER commands from ISQL or REXX

EXECs. You must have DB2 REXX SQL for VM/ESA (RXSQL) installed

before you can use REXX with DB2 Server for VM.

COUNTER POOL

Displays the number of occurrences of key events by storage pool. Also displays

counters for the directory and unmapped internal dbspaces.

Syntax

►► COUNTER POOL

►◄

DIR

UNMAPPED

Displays counters for all storage pools, the directory, and unmapped

internal dbspaces.

Displays counters for storage pool number n. Valid values are integers

from 1 to 999.

DIR Displays counters for the directory.

UNMAPPED

Displays counters for unmapped internal dbspaces.

Description

Storage Pool Counters: The COUNTER POOL command displays two sets of

counters. One set is for storage pools using Data Spaces Support, a directory that

uses Data Spaces Support, and unmapped internal dbspaces. The other set is for

pools using the standard DASD I/O system and a directory using the standard

system.

The counters for the standard I/O system are:

LBUFLOOK

The number of times the database manager has looked into the local data

page buffers.

PGREAD

The number of pages moved from DASD to the local buffers.

PGWRITE

The number of pages moved from the local buffers to DASD.

199

IUCVBIO

The number of IUCV *BLOCKIO requests submitted by the database

manager to the operating system. Note that, because the database manager

often uses one IUCV *BLOCKIO request to write several pages at one time,

IUCVBIO is usually lower than PGREAD+PGWRITE.

The counters for Data Spaces Support are:

LBUFLOOK

The number of times the database manager has looked into the local data

page buffers.

DSREAD

The number of pages moved from the VM data spaces to the local buffers.

DSWRITE

The number of pages moved from the local buffers to the VM data spaces.

DSFAULT

The number of pages, requested from data spaces, that the operating

system did not find in main or expanded storage. These page faults cause

the operating system to read the pages from DASD in to the data spaces in

main storage.

Example

This example shows a database where the directory and the first storage pool are

using Data Spaces Support, the second pool is using the standard DASD I/O

system, and the internal dbspaces are mapped. Because the internal dbspaces are

included in storage pool 1, any activity recorded for them will be included with

the totals for pool number 1.

200

Operation

counter pool *

Counter values at DATE=’05-24-91’ TIME=’10:13:49’

Directory: Data Spaces

Pages looked at in the buffer LBUFLOOK:

Pages moved from DS to buffer DSREAD

Pages moved from buffer to DS DSWRITE :

DS page fault notifications

DSFAULT :

Pool No. 1: Data Spaces

Pages looked at in the buffer LBUFLOOK:

1005

Pages moved from DS to buffer DSREAD

Pages moved from buffer to DS DSWRITE :

DS page fault notifications

DSFAULT :

Pool No. 2:

*BLOCKIO

Pages looked at in the buffer LBUFLOOK:

237

Page reads

PGREAD

Page writes

PGWRITE :

IUCV *BLOCKIO I/O requests

IUCVBIO :

Total count for storage pools using data spaces:

Pages looked at in the buffer LBUFLOOK:

1005

Pages moved from DS to buffer DSREAD

Pages moved from buffer to DS DSWRITE :

DS page fault notifications

DSFAULT :

Total count for storage pools using *BLOCKIO:

Pages looked at in the buffer LBUFLOOK:

237

Page reads

PGREAD

Page writes

PGWRITE :

IUCV *BLOCKIO I/O requests

IUCVBIO :

ARI0065I Operator command processing is complete.

This example shows a database where the directory is using the standard DASD

I/O system.

counter pool dir

Counter values at DATE=’05-23-91’ TIME=’15:41:07’

Directory:

*BLOCKIO

Pages looked at in the buffer LBUFLOOK:

410

Page reads

PGREAD

Page writes

PGWRITE :

IUCV *BLOCKIO I/O requests

IUCVBIO :

ARI0065I Operator command processing is complete.

This example shows a database using unmapped internal dbspaces. Any activity

recorded for these dbspaces will only appear under the heading Unmapped Internal

Dbspaces. It will not be included in the totals for the storage pool that contains the

internal dbspaces.

Appendix B. VMDSS Operator Commands

201

counter pool unmapped

Counter values at DATE=’07-29-91’ TIME=’17:12:37’

Unmapped Internal Dbspaces: Data Spaces

Pages looked at in the buffer LBUFLOOK:

515

Pages moved from DS to buffer DSREAD

Pages moved from buffer to DS DSWRITE :

DS page fault notifications

DSFAULT :

ARI0065I Operator command processing is complete.

RESET POOL

Resets storage pool counters to zero. Also resets the counters for the directory and

unmapped internal dbspaces to zero.

Syntax

►► RESET POOL

►◄

DIR

UNMAPPED

Resets the counters for all storage pools, the directory, and unmapped

internal dbspaces to zero.

Resets the counters for storage pool number n to zero. Valid values are

integers from 1 to 999.

DIR Resets the counters for the directory to zero.

UNMAPPED

Resets counters for unmapped internal dbspaces.

Description

This operator command resets storage pool and directory counters to zero. You can

use this command to develop a picture of database performance over time. At

regular intervals, display the storage pool and directory counters with the

COUNTER POOL command; then reset the counters with RESET POOL. The next

time you display the counters, they will display the number of key events that

occurred between the time when you issued the RESET POOL command and the

time when you issued COUNTER POOL.

202

Operation

Example

counter pool 1

Counter values at DATE=’05-24-91’ TIME=’10:11:07’

Pool No. 1: Data Spaces

Pages looked at in the buffer LBUFLOOK:

1005

Pages moved from DS to buffer DSREAD

Pages moved from buffer to DS DSWRITE :

DS page fault notifications

DSFAULT :

ARI0065I Operator command processing is complete.

reset pool 1

Counters reset at DATE=’05-24-91’ TIME=’10:11:15’

ARI0065I Operator command processing is complete.

counter pool 1

Counter values at DATE=’05-24-91’ TIME=’10:11:22’

Pool No. 1: Data Spaces

Pages looked at in the buffer LBUFLOOK:

Pages moved from DS to buffer DSREAD

Pages moved from buffer to DS DSWRITE :

DS page fault notifications

DSFAULT :

ARI0065I Operator command processing is complete.

SET POOL

Changes storage pool specifications.

Syntax

(1)

►► SET POOL n1

►◄

-n2

STR

SEQ

Notes:

You must include at least one of these blocks.

Specifies that you want to change the specifications for storage pool n1.

Valid values are integers from 1 to 999.

Specifies a range of storage pools from n1 to n2. Valid values are integers

from 1 to 999. n2 must be greater than or equal to n1.

DS Sets the working storage residency priority of the storage pools you specify

to n. Valid values are integers from 1 to 5. The default is 3. If you do not

include DS the database manager will use the current storage pool setting

(unless the pools you specified are not using data spaces).

Table 3 on page 204 summarizes the five storage residence priorities. An R

indicates that the database manager releases a page from main and

expanded storage after it has been moved to a local buffer.

Appendix B. VMDSS Operator Commands

203

Table 3. Storage Residence Priorities

Current Working

Page Type

Storage Size

Working Storage Residence Priority

Data

≤ target

> target

Index

≤ target

> target

STR Turns striping on for the storage pools you specify.

SEQ Turns striping off for the storage pools you specify. The database manager

will allocate pages sequentially on DASD.

Note: If you do not include SEQ or STR, the database manager will use

the current storage pool setting.

Description

This command lets you dynamically change some of the current storage pool

specifications. Note that since you cannot use this command to dynamically change

whether a storage pool uses Data Spaces Support, the storage pool you specify

must already be using data spaces before you can use the DS parameter.

Example

The following example shows a summary of storage pools before and after turning

striping on for storage pool 2:

show pool summary

POOL

TOTAL NO. OF

NO. OF

% NO. OF

NO.

PAGES PAGES USED FREE PAGES RESV PAGES USED EXTENTS SOS I/O

1710

122

1588

D2S

7467

7465

FREE

1674279

ARI0065I Operator command processing is complete.

set pool 2 str

ARI0065I Operator command processing is complete.

show pool summary

POOL

TOTAL NO. OF

NO. OF

% NO. OF

NO.

PAGES PAGES USED FREE PAGES RESV PAGES USED EXTENTS SOS I/O

1710

122

1588

D2S

7467

7465

BKS

FREE

1674279

ARI0065I Operator command processing is complete.

This example shows storage pool 1 before and after turning striping off and setting

its working storage residence priority to 2:

204

Operation

show pool 1

POOL NO.

NUMBER OF EXTENTS = 2

DS3 STR

EXTENT TOTAL NO. OF

NO. OF

NO.

PAGES PAGES USED FREE PAGES RESV PAGES USED

855

781

855

808

TOTAL

1710

121

1589

ARI0065I Operator command processing is complete.

set pool 1 ds2 seq

ARI0065I Operator command processing is complete.

show pool 1

POOL NO.

NUMBER OF EXTENTS = 2

DS2 SEQ

EXTENT TOTAL NO. OF

NO. OF

NO.

PAGES PAGES USED FREE PAGES RESV PAGES USED

855

781

855

808

TOTAL

1710

121

1589

ARI0065I Operator command processing is complete.

SET SAVEINTV

Sets the number of modified pages the database manager allows in data spaces

before it writes them to DASD.

Syntax

►► SET SAVEINTV n

►◄

Sets the save interval for the database to n blocks of 4KB pages (32 pages

in a block). When the number of modified blocks of pages in a data space

exceeds this parameter, the database manager directs the operating system

to save all the pages in that data space that were modified by the database

manager to DASD. Valid values are integers from 1 to 999,999 blocks. The

default is 10 blocks.

Description

This command lets you dynamically change the VMDSS save interval. It also

displays your previous save interval and your new one.

Example

set saveintv 12

Previous SAVEINTV = 10

SAVEINTV = 12

ARI0065I Operator command processing is complete.

Note: This command is not allowed when no storage pools are using data spaces.

For example, when the storage pool specification file specifies BLK for all

Appendix B. VMDSS Operator Commands

205

storage pools. This is true even if the directory is in a data space, or if

unmapped data spaces are used for internal dbspaces (if SEPINTDB=Y is

used), or both.

SET TARGETWS

Sets the target working storage size of the database.

Syntax

►► SET TARGETWS n

►◄

Sets the target working storage size for the database to n megabytes. This

limits the amount of main and expanded storage that the database machine

will use. While the database machine may exceed this target, the database

manager will to keep you at or below it if possible. Valid values are

integers from 1 to 999,999 megabytes. The default is 32 megabytes.

Description

This command lets you dynamically change the target working storage size. It also

displays the following information about the database’s:

v Previous target working storage size

v Current target working storage size

v Current working storage size

v How much of the current working storage is in main storage and how much is

in expanded storage.

Example

set targetws 30

Previous TARGETWS = 32 MB

TARGETWS = 30 MB

Current Working Storage = 5 MB

(5 MAIN + 0 EXPANDED)

ARI0065I Operator command processing is complete.

SHOW ACTIVE

Displays the status of active database users.

Syntax

►► SHOW ACTIVE

►◄

Description

This command shows you whether a user agent is waiting for a data space page

fault or the checkpoint agent is waiting for an SLD block. If a user agent is waiting

for a data space page fault, you will see:

Agent ... is in DS page fault Wait

206

Operation

The checkpoint agent is waiting for an SLD block, you will see:

Checkpoint agent is in DS SLD block Wait

Example

show active

Status of SQL/DS agents:

Checkpoint agent is not active

User Agent:

1 User-ID: SMITH

is R/W APPL

7B4

Agent is

processing and is in I/O

Wait

User Agent:

2 User-ID: MICHAEL is R/O SUBS

7B9

Agent is not processing and is in communication Wait

User Agent:

3 User-ID: JESSICA is R/O APPL

5A4

Agent is

processing and is in DS page fault Wait

User Agent:

4 User-ID: TESTUSER is R/W APPL

7BB

Agent is

processing and is in DS page fault Wait

agent(s) not connected to an APPL or SUBSYS

ARI0065I Operator command processing complete

show active

Status of SQL/DS agents:

Checkpoint agent is in DS SLD block Wait

User Agent:

1 User-ID: PETER

is NEW APPL

106

Agent is waiting for log archive checkpoint.

Wait

User Agent:

2 User-ID: GENE

is NEW SUBS

107

Agent is waiting for log archive checkpoint.

agent(s) not connected to an APPL or SUBSYS

ARI0065I Operator command processing complete

SHOW BUFFERS

Displays information about the buffer pool size and usage.

Syntax

►► SHOW BUFFERS

►◄

Description

This command shows you whether a specific local buffer or directory buffer

contains a page that comes from a data space. If a buffer contains a data space

page, the FLAGS column will contain a hex value where the X'10' bit (00010000) is

turned on.

To determine whether a particular page is from a data space, either check if the

most significant digit of the flag is an odd number, or perform a logical AND

between the FLAGS byte and X'10'. If the digit is odd, or if the result of the

calculation is X'10', the page is from a data space.

For example, a FLAGS byte of X'D0':

In Hex:

’D0’ & ’10’ = ’10’

In Binary:

11010000 & 00010000 = 00010000

Appendix B. VMDSS Operator Commands

207

Also the first digit of ’D0’ (’D’ or 13) is an odd number. This page is from a data

space.

The FLAGS byte contains four flag bits in total:

’80’

Modified Page. The buffer page has been modified.

’40’

In Transit The buffer page is being copied to or from DASD or a data

space.

’20’

New Page The buffer page is not from DASD or a data space; the buffer

contains a new page.

’10’

Data Space Page The buffer page is from a data space.

The FLAGS byte does not use the four low order bits. They will always be zero.

Example

Note that the hex values under the FLAGS column are cumulative. For example,

the directory buffer at address 0047A000 contains a modified (identified by X'80')

data space page (identified by X'10').

show buffers

Only used buffers are displayed.

DBSPACE REC

ADDR

FLAGS FIX CNT

Page Buffers

0006FC

0045C000

32001

000093

0045D000

000080

0045E000

00009D

0045F000

00009C

00478000

00009B

00479000

Directory Buffers

000001

0047A000

000002

0047A200

00000D

0047A400

000005

0047C800

000001

0047CA00

ARI0065I Operator command processing is complete.

SHOW LOCK ACTIVE

Displays locking information for active database users.

Syntax

►► SHOW LOCK ACTIVE

►◄

Description

This command shows you whether a user agent is waiting for a data space page

fault. If it is waiting for a data space page fault, you will see DSPF in the WAIT

STATE column.

208

Operation

Example

show lock active

WAIT TOTAL LONG WANTLOCK WANTLOCK

AGENT USER

STATE LOCKS LOCKS TYPE

DBSPACE

JESSICA I/O

140

100

LEIGH

LOCK

226

220

DBSP

MARY

COMM

MIKE

LOCK

108

101

DBSP

BOB

LOCK

107

PAGE

GENE

DSPF

EDWIN

DSPF

101

ARI0065I Operator command processing complete

SHOW LOCK GRAPH

Displays information about the lock dependencies between active database users.

Syntax

►► SHOW LOCK GRAPH

►◄

Description

This command shows you whether a user agent is waiting for a data space page

fault. If it is waiting for a data space page fault, you will see DSPF in the WAIT

STAT column.

Example

show lock graph

LOCK

WAIT LOCK DBSP LOCKK

REQ

REQUESTER

HOLDER

STAT TYPE NUMBR QUALFIER

STATE

MODE DUR

KEVIN

MARK DSPF DBSP

G WAIT S

LONG

MARK

MIKE LOCK DBSP

G WAIT X

LONG

MIKE

LAURA LOCK PAGE

G WAIT IS

LONG

LAURA

STEVE COMM DBSP

G WAIT IX

LONG

LAURA

GRANT DSPF DBSP

G WAIT IX

LONG

ARI0065I Operator command processing complete

SHOW POOL

Displays physical storage information about database storage pools.

Syntax

ALL

►► SHOW POOL

►◄

SUMMARY

DELETED

Appendix B. VMDSS Operator Commands

209

ALL Displays detailed information, grouped by dbextent, about each storage

pool in the database.

SUMMARY

Displays summary information about each storage pool in the database.

DELETED

Display information about deleted dbextents

Displays detailed information, grouped by dbextent, about storage pool

number n.

Description

This command is updated to let you display the following for storage pool

information:

v Whether it is using data spaces

v Which working storage residency priority it is using

v Whether it is using striping.

SHOW POOL SUMMARY: A column entitled “I/O” is added to the SHOW

POOL SUMMARY display. Under its heading is a two-or three-letter code that

indicates the current storage pool specifications. Possible specification code

combinations are:

BK The storage pool is using the standard I/O system, without striping.

BKS The storage pool is using the standard I/O system, with striping.

Dn The storage pool is using Data Spaces Support, without striping. n is the

current storage residency priority.

DnS The storage pool is using Data Spaces Support, with striping. n is the

current storage residency priority.

SHOW POOL and SHOW POOL ALL: Two new codes are added to the SHOW

POOL and the SHOW POOL ALL displays beside the number of extents. Possible

specification code combinations are:

BLK SEQ

The storage pool is using the standard I/O system, without striping.

BLK STR

The storage pool is using the standard I/O system, with striping.

DSn SEQ

The storage pool is using Data Spaces Support, without striping. n is the

current storage residency priority.

DSn STR

The storage pool is using Data Spaces Support, with striping. n is the

current storage residency priority.

210

Operation

Example

show pool 1

POOL NO.

NUMBER OF EXTENTS = 2

DS2 STR

EXTENT TOTAL NO. OF

NO. OF

NO.

PAGES PAGES USED FREE PAGES RESV PAGES USED

855

778

855

810

TOTAL

1710

122

1588

ARI0065I Operator command processing is complete.

show pool all

POOL NO.

NUMBER OF EXTENTS = 2

DS2 STR

EXTENT TOTAL NO. OF

NO. OF

NO.

PAGES PAGES USED FREE PAGES RESV PAGES USED

855

778

855

810

TOTAL

1710

122

1588

POOL NO.

NUMBER OF EXTENTS = 1

BLK STR

EXTENT TOTAL NO. OF

NO. OF

NO.

PAGES PAGES USED FREE PAGES RESV PAGES USED

7467

7465

TOTAL

7467

7465

FREE AREAS:

NUMBER OF DELETED EXTENTS = 0

EXTENT TOTAL

NO.

PAGES

END

1674279

TOTAL

1674279

Maximum number of DBEXTENTs = 6

ARI0065I Operator command processing is complete.

show pool summary

POOL

TOTAL NO. OF

NO. OF

NO.

PAGES PAGES USED FREE PAGES RESV PAGES USED

EXTENTS SOS I/O

1710

122

1588

D2S

7467

7465

BKS

FREE

1674279

ARI0065I Operator command processing is complete.

SHOW SAVEINTV

Displays the current number of modified blocks of pages the database manager

allows in data spaces before it writes them to DASD.

Syntax

►► SHOW SAVEINTV

►◄

Appendix B. VMDSS Operator Commands

211

Description

This command displays the database’s current save interval. The save interval is

measured in blocks of 4KB pages (32 pages in a block). The save interval is the

upper limit on the number of modified blocks allowed in a data space at any one

time. When the number of modified blocks in any one data space exceeds this

limit, the database manager writes all the modified blocks in that data space to

DASD. The default is 10 blocks (or 320 pages).

Note: For a block to be “modified” the database manager only needs to change

one page in that block. The block will remain “modified” until it is written

to DASD.

Note: This command is not allowed when no storage pools are using data spaces.

For example, when the storage pool specification file specifies BLK for all

storage pools. This is true even if the directory is in a data space, or if

unmapped data spaces are used for internal dbspaces (if SEPINTDB=Y is

used), or both.

Example

show saveintv

SAVEINTV = 10

ARI0065I Operator command processing is complete.

SHOW TARGETWS

Displays the current target working storage size of the database.

Syntax

►► SHOW TARGETWS

►◄

Description

This command displays the following information about the database’s:

v Its current target working storage size

v Its current working storage size

v How much of the current working storage is in main storage and how much is

in expanded storage.

The target working storage size and the working storage size are measured in

megabytes. The former is the amount of main and expanded storage that the

database manager will try to use for the database. The default is 32 megabytes.

Example

show targetws

TARGETWS = 32 MB

Current Working Storage = 5 MB

(5 MAIN + 0 EXPANDED)

ARI0065I Operator command processing is complete.

212

Operation

Internal Counters

VMDSS contains a set of internal counters that are primarily intended for IBM’s

use. However, if you have a good understanding of the CP Macros that control

data spaces within VM/ESA, you may find the following counters useful.

ESASEG

The number of data space segments mapped. Data spaces are divided into

256-page chunks called segments. A segment is mapped the first time the

database manager references a dbspace page that lies in that segment (the

first time it references a dbspace, the database manager decides where in

the data space the page will be). Mapping a segment consists of defining a

mapping between every dbspace page in that segment and its place on

disk. Empty pages are not mapped (because there is no disk page allocated

for an empty page). Even if all the dbspace pages in a data space segment

are empty, the database manager still marks that segment as mapped and

increments ESASEG.

MAPSEGPG

The number of calls to CP’s MAPMDISK DEFINE macro to map the above

ESASEG segments. It may take 0 (all pages empty) or many DEFINEs per

segment, because one DEFINE can only work on a contiguous set of data

space pages and database manager must skip empty pages.

REQMAPPG

The number of requests to remap a page. When the database manager gets

a new disk page for a dbspace page (either when database manager

shadows it or starts using a previously empty dbspace page), it must

redefine the data space mapping for that page to point to the new disk

location.

MAPPG

The number of MAPMDISK DEFINE macro calls done to satisfy the

REQMAPPG requests above. The database manager tries to delay the

requests as long as possible in order to group as many as it can into one

DEFINE. In the worst case, there will be 1 DEFINE for every request. In

the best case, 1 for every 256 requests.

SAVESLD

The number of calls to CP’s MAPMDISK SAVE macro to save the contents

of a page in the Data Space on DASD.

SAVEMXRN

The sum of “maximum runs”. Before issuing a MAPMDISK SAVE macro

for one or more blocks of 32 pages, a maximum run is computed. It is the

largest number of contiguous blocks of 32 pages that are being saved.

SAVEMXRN is an indicator of whether the blocks of pages being saved are

consecutively modified (during a data load for example) and determines

the span length to use for the REFPAGE macro following completion of

MAPMDISK SAVE macro (refer to SCBLKREF and SCSPNREF).

SAVEBLK

The number of 32 page blocks for which CP’s MAPMDISK SAVE macro

was issued. The MAPMDISK SAVE macro is issued for modified blocks of

pages at checkpoints and when the number of modified blocks reaches the

SAVEINTV threshold.

WAITSLD

The number of times the database manager was forced to wait for a

MAPMDISK SAVE macro to save modified pages on DASD. To ensure that

Appendix B. VMDSS Operator Commands

213

all modified data is safely stored on DASD, the checkpoint processing

cannot finish until all outstanding MAPMDISK SAVE requests are

complete.

REFBLOCK

The number of times a block form of CP’s REFPAGE macro was issued

telling CP the reference pattern for a group of pages.

REFBPAGE

The number of pages for which a block form of REFPAGE macro was

issued.

REFBSPAN

The sum of spans used in block form of REFPAGE macro.

REFLST

The number of times a list form of CP’s REFPAGE macro was issued.

REFLPAGE

The number of pages for which a list form of REFPAGE macro was issued.

DIAG10

The number of DIAGNOSE X’10’ instructions processed that release the

main or expanded storage back to CP.

DIAG10PG

The number of data space pages that have been released using the

DIAGNOSE X’10’ instruction.

SCDIAG10

The number of blocks of 32 pages for which storage has been released

following the completion of MAPMDISK SAVE request.

SCREF

The number of blocks of 32 pages for which a reference pattern was

established following the completion of MAPMDISK SAVE request.

SCBLKREF

The number of times a block form of the REFPAGE macro was issued

following the completion of MAPMDISK SAVE request.

SCSPNREF

The sum of spans used in the block form of the REFPAGE macro following

the completion of the MAPMDISK SAVE request. The size of each span is

based on the largest number of consecutive blocks of 32 pages that were

(SAVEMXRN counter). As the number of consecutively modified blocks

increases (during a data load for example) the size of the span given to

CP’s REFPAGE macro also increases.

SAVEGNRL

The number of save requests made by general agents when the SAVEINTV

has been reached for a data space.

SAVECHK0

The number of save requests made by the checkpoint agent to flush the

page buffers prior to starting checkpoint processing.

SAVECHK1

The number of save requests made by the checkpoint agent during

checkpoint processing.

To display the counters, type the following at the operator console:

COUNTER INTERNAL *

214

Operation

For example:

counter internal *

Counter values at DATE=’07-23-91’ TIME=’08:57:51’

ESASEG

429

MAPSEGPG 432

REQMAPPG 15

MAPPG

SAVESLD

SAVEMXRN 1

SAVEBLK

WAITSLD

REFBLOCK 860

REFBPAGE 163912

REFBSPAN 68485

REFLST

1580

REFLPAGE 101120

DIAG10

1580

DIAG10PG 101120

SCDIAG10 0

SCREF

SCBLKREF 0

SCSPNREF 0

SAVEGNRL 0

SAVECHK0 0

SAVECHK1 0

ARI0065I Operator command processing is complete.

To reset the counters to zero, type the following:

RESET INTERNAL *

For example:

reset internal *

Counters reset at

DATE=’05-23-91’ TIME=’09:26:15’.

counter internal *

Counter values at

DATE=’07-23-91’ TIME=’09:26:51’

ESASEG

MAPSEGPG 0

REQMAPPG 0

MAPPG

SAVESLD

SAVEMXRN 0

SAVEBLK

WAITSLD

REFBLOCK 0

REFBPAGE 0

REFBSPAN 0

REFLST

REFLPAGE 0

DIAG10

DIAG10PG 0

SCDIAG10 0

SCREF

SCBLKREF 0

SCSPNREF 0

SAVEGNRL 0

SAVECHK0 0

SAVECHK1 0

ARI0065I Operator command processing is complete.

Appendix B. VMDSS Operator Commands

215

DB2 Server for VSE. Operations Guide / Handbooks (2004-2007) - page 26