Setting Application Server Domain Parameters

This chapter describes all of the configuration options that are related to an application server domain. Generally, the documentation reflects the order in which the configuration sections appear in the PSADMIN interface or the PSAPPSRV.CFG file.

This chapter discusses:

Note. As a configuration option, you can configure a domain to spawn server processes according to the volume of transaction requests. There is no explicit parameter that you must set to enable spawning. In the following configuration section descriptions, some servers enable you to specify a minimum and maximum number of server processes. To enable spawning for these server processes, the maximum value must exceed the minimum value by an increment of at least one. As needed, the domain spawns server processes up to the maximum value. As the volume of transactions decreases, the number of spawned server processes decreases, or decays, until the minimum value is reached. By default, spawning is disabled.

Click to jump to parent topicStartup Options

Set database sign-in values in the Startup section, so that the domain can connect to the database.

Note. When specifying passwords, such as UserPswd, Connect Password, and StandbyUserPswd you will be prompted with: Do you want to encrypt this password? Selecting y encrypts the password so that it does not appear in clear text in the configuration file (PSAPPSRV.CFG) or PSADMIN.

Click to jump to top of pageClick to jump to parent topicDBName

Enter the PeopleSoft database name, such as FSDMO or HRDMO. This parameter is case sensitive.

Click to jump to top of pageClick to jump to parent topicDBType

Enter the PeopleSoft database type, such as DB2ODBC, DB2UNIX, INFORMIX, MICROSFT, ORACLE, or SYBASE. If you enter an invalid database type, PSADMIN prompts you with a valid list.

Click to jump to top of pageClick to jump to parent topicUserID

Enter the PeopleSoft user ID that is authorized to start the application server. For the application server to boot, the appropriate user ID with the correct authorizations must be assigned to this parameter. This is the ID that the application server passes to the database for authentication and connection. The user ID that you enter here is not related to the actual user (administrator) that carries out the boot command.

The Can Start Application Server permission must be set in the permission list. The authorization to start an application server does not (directly or indirectly) grant any authorizations or privileges beyond the ability to start the application server. Each user who attempts to sign in enters a unique user ID and password, which the application server uses to authenticate each user.

See Also

Setting General Permissions

Click to jump to top of pageClick to jump to parent topicUserPswd

Enter the password to be used by the specified user ID that will gain access to the database.

The password:

Click to jump to top of pageClick to jump to parent topicConnect ID

Required for all database platforms. Enter the database-level ID that the PeopleSoft system uses to make the initial connection to the database. This user name must have authority to select from PSACCESPRFL, PSLOCK, PSOPRDEFN, and PSSTATUS.

Click to jump to top of pageClick to jump to parent topicConnect Password

Enter the password for the connect ID. For instance, this might be the UNIX user's password (either uppercase or lowercase).

The password:

Click to jump to top of pageClick to jump to parent topicServerName

Required for Sybase and Informix. Enter the name of the server on which the PeopleSoft database is installed. This value is case sensitive.

Click to jump to top of pageClick to jump to parent topicStandbyDBName

(Used only for by Oracle databases with Oracle Active Data Guard implemented). Specify the standby database name.

Click to jump to top of pageClick to jump to parent topicStandbyDBType

(Used only for by Oracle databases with Oracle Active Data Guard implemented). Specify the standby database type, such as ORACLE.

Click to jump to top of pageClick to jump to parent topicStandbyUserId

(Used only for by Oracle databases with Oracle Active Data Guard implemented). Specify the user ID required to connect to the standby database.

Click to jump to top of pageClick to jump to parent topicStandbyUserPswd

(Used only for by Oracle databases with Oracle Active Data Guard implemented). Specify the user password required to connect to the standby database.

Click to jump to parent topicDatabase Options

Use the Database Options section to specify environment variables that may improve the performance of the system. These options do not apply to every database.

Click to jump to top of pageClick to jump to parent topicSybasePacketSize

Enter a Transmission Control Protocol (TCP) packet size. The minimum value is 512 and the maximum value is 65538. The default packet size is 512. If you change the packet size, make the corresponding changes to the Sybase database server.

See Your Sybase documentation.

Click to jump to top of pageClick to jump to parent topicUseLocalOracleDB

(Applies only to Oracle databases.). Use this option to enable processes to initiate a local connection to a PeopleSoft database that is running on the same machine. You should use this option for all PeopleSoft Process Scheduler (batch) and application server configurations that are local (on the same server) to the PeopleSoft Oracle instance. This type of connection enables processing to complete significantly quicker.

Enter 1 to enable this option, and enter 0 to disable it.

Note. Using the local Oracle connection disables the Query Kill function.

Note. On Microsoft Windows, UseLocalOracleDB is not supported when using a 32-bit client connecting to a 64-bit database.

See Also

PeopleSoft Servers and the Oracle Connection String

Click to jump to top of pageClick to jump to parent topicEnableDBMonitoring

Required for database-level auditing. How this works varies slightly, depending on the database platform. Use this option to view more information regarding the clients that are connected to a database server through the application server. For instance, with this enabled, you can view the client machine name or user ID that is associated with a particular connection. Without this option enabled, all connections appear somewhat anonymously, as in PSFT or APPSERV.

The default value is 1 (enabled). Enter 0 to disable it.

Note. This parameter isn't supported for Informix or DB2 LUW.

Click to jump to top of pageClick to jump to parent topicPSDB Maximum Cursors

Enables you to configure the maximum number of cursors opened at one time. In general, the maximum number of cursors that can be open at one time within PeopleTools is 64,000. However, some database platforms place greater restrictions on the number of cursors that can be open at one time. By default, if no value is provided, the system allows a maximum of 1024 cursors.

To increase or decrease the default value, assign your custom value to this parameter. The value you set should be determined on multiple factors including:

Always refer to your database documentation for the latest recommendations related to maximum cursors settings. The following information can be used as a guideline.

Database

Maximum Cursors Limit

Oracle

No predefined limit.

IBM DB2 z/OS

Limited only by available storage.

IBM DB2 LUW

Limited only by available storage.

Microsoft SQL Server

No predefined limit.

Sybase

No predefined limit.

Informix

No predefined limit, but SQL statement length limited to 64 KB.

Click to jump to parent topicSecurity Options

Use the Security section to set an additional layer to the sign-in process.

Click to jump to top of pageClick to jump to parent topicValidate Signon With Database

Use this option to set an additional level of authorization-checking to be performed at the database level. Enter 1 to enable this option, and enter 0 to disable it.

With this option disabled, if a PeopleSoft user attempts to connect to an application server, the application server ensures that the user's PeopleSoft user ID and password exist on PSOPRDEFN. If it does not exist, the request to connect fails. This is PeopleTools-level authentication.

With this option enabled, the application server first attempts to connect to the database by using the user ID and password as part of the database connection string. If the authorization is successful, it disconnects, and then the normal PeopleSoft sign-in procedure occurs.

With this option enabled, to connect successfully to the database, the user must be defined on either the operating system or the database and within PeopleSoft.

Note. For DB2 z/OS (MVS), the user ID and password must be defined as z/OS user logon IDs.

Click to jump to top of pageClick to jump to parent topicDomainConnectionPwd

The domain connection password adds an extra layer of security between the application server domain and any connections made to it. This password enables you to further prevent unauthorized clients from establishing connections to an application server domain. It is recommended to use PSADMIN to update this value.

All domains, PeopleSoft Internet Architecture, and three-tier workstations used for a particular database, must use the same domain connection password. The default value is PS.

Note. If you change this value in the domain configuration, you must also update any PeopleSoft Internet Architecture sites and three-tier Windows workstations. If no value is specified, the system assumes the default of PS.

For the PeopleSoft Internet Architecture configuration, you enter the password in the configuration.properties file using the same setting, DomainConnectionPwd.

For a three-tier Windows workstation connection, you enter the password in the Configuration Manager profile using the Domain Connection Password field on the Database/Application Server tab of the Edit Profile dialog box.

Note. If signing on with a three-tier workstation without first configuring the workstation with Configuration Manager (as in specifying the domain and WLS port in the Application Server Name field), the signon attempt submits PS as the domain connection password.

See Also

Configuring Domain Connection Password

Domain Connection Password

Click to jump to parent topicWorkstation Listener Options

The workstation listener is the component to which PeopleSoft development environments running on Windows (Application Designer, for example) send Tuxedo messages. By default, the workstation listener is disabled.

Click to jump to top of pageClick to jump to parent topicAddress

%PS_MACH% resolves automatically to the machine name that PSADMIN obtains by using a system application programming interface (API) call. You can also specify the machine's Internet Protocol (IP) address (dotted notation) or its resolvable name (domain name server [DNS] name).

You should not change this value except in the following rare cases. If you are configuring files to run an application server on another machine (that is, you plan to copy PSAPPSRV.CFG and PSAPPSRV.UBB to a domain on another machine), you must overlay %PS_MACH% with the other machine's name.

Note. If you enter a literal IP address or machine name in place of the %PS_MACH% system variable, PSADMIN automatically prepends '//' to the value during the configure process. For example, if you enter 10.831.248.117 in place of %PS_MACH%, after configuring the domain, the value appears as //10.831.248.117 in both PSADMIN and the psappsrv.cfg file.

Click to jump to top of pageClick to jump to parent topicPort

Enter the 4-digit port number to assign to the WSL. Port numbers are arbitrary numbers between 1000 and 64 K and must not already be in use by another service. The default value is 7000.

Click to jump to top of pageClick to jump to parent topicEncryption

Use this option to enable the encryption of data messages between client workstations and the application server. Specify one of the following values:

Note. Because this is a dynamic parameter, you must modify it by selecting Custom Configuration on the Quick-Configure menu, and reboot the application server domain for it to take effect.

Click to jump to top of pageClick to jump to parent topicMin Handlers

Enter the number of workstation handlers (WSHs) to be started at boot time. The default for small and large application server configuration templates are 1 and 10, respectively.

Click to jump to top of pageClick to jump to parent topicMax Handlers

Enter the maximum number of WSHs that can be started for a domain. If the Min Handlers value equals the Max Handlers value, Tuxedo does not automatically spawn incremental WSHs.

Click to jump to top of pageClick to jump to parent topicMax Clients per Handler

Enter the maximum number of client workstation connections that each WSH can manage. Each WSH can handle approximately 60 client connections. Numbers vary depending upon the resources of the server. In most cases, you should decrease the default as opposed to increasing it. The default is 40.

Click to jump to top of pageClick to jump to parent topicClient Cleanup Timeout

Enter the amount of time, in minutes, that a client connection can remain idle (no work requested) before Tuxedo terminates the client connection. Client disconnects are transparent to a client, and a user just clicks the mouse to cause a reconnection. The default value for this setting is 60 minutes.

Click to jump to top of pageClick to jump to parent topicInit Timeout

This value, when multiplied by SCANUNIT (a UBB parameter value that is defined in the PSAPPSRV.UBB file) specifies the amount of time, in seconds, that Tuxedo allows for a client connection request to bind to a WSH before terminating the connection attempt.

Click to jump to top of pageClick to jump to parent topicTuxedo Compression

Enter the minimum length of a data message for which the application server initiates data compression. While compression results in favorable performance gains for transactions over a wide area network (WAN), testing reveals that compression can degrade performance slightly over a local area network (LAN) due to the compression and decompression overhead.

You should use the default threshold of 5000, which sets a balance between WAN and LAN environments. This means that only network request and response messages over 5000 bytes are compressed, and those 5000 and under are uncompressed. If you support both WAN and LAN users, you can configure a hybrid environment by configuring two application servers: one to support WAN users (with compression set to 100) and another to support LAN users (with compression set to 100000, effectively turning compression off).

Click to jump to parent topicJolt Listener Options

Use this section to configure PeopleSoft Internet Architecture connections. The Jolt listener enables Tuxedo to exchange messages with the web server.

Click to jump to top of pageClick to jump to parent topicAddress

See the equivalent parameter for the workstation listener.

Click to jump to top of pageClick to jump to parent topicPort

Enter the port number that is used for the Jolt server listener (JSL). This value can be any port number that is not already in use by another service on the machine that runs the application server domain. By default, the JSL port is enabled.

Click to jump to top of pageClick to jump to parent topicEncryption

Use this option to enable the encryption of data messages between client workstations and the web server. Specify one of the following values:

Click to jump to top of pageClick to jump to parent topicMin Handlers

Enter the number of Jolt server handlers (JSH) to be started at boot time. Each JSH multiplexes up to 50 connections.

Click to jump to top of pageClick to jump to parent topicMax Handlers

Enter the maximum number of JSHs.

Note. JSHs spawn by using successive port numbers starting at the port number for the JSL in the PSAPPSRV.CFG file. Make sure that the additional ports are free before configuring spawning.

Click to jump to top of pageClick to jump to parent topicMax Clients per Handler

Enter the maximum number of client connections that each JSH can manage.

Click to jump to top of pageClick to jump to parent topicClient Cleanup Timeout

Enter the amount of time, in minutes, that a client connection can remain idle (no work requested) before Tuxedo terminates the client connection. Client disconnects are transparent to a client, and a user just clicks the mouse to cause a reconnection. The default value for this setting is 10 minutes.

Click to jump to top of pageClick to jump to parent topicInit Timeout

See the equivalent parameter for the workstation listener.

Click to jump to top of pageClick to jump to parent topicClient Connection Mode

Enter one of these options to control the allowed connection modes from clients:

Click to jump to top of pageClick to jump to parent topicJolt Compression Threshold

Jolt compression can significantly improve performance. Jolt compression enables messages that are transmitted through a Jolt connection to be compressed as they flow over the network. You are likely to see the most significant performance improvements over a WAN.

For compression, the configuration files contain a default compression threshold. This default value should provide the best results for most situations. However, your application server administrator can adjust this value to suit your implementation.

The compression threshold indicates to the server how large a packet must be to require compressing. In other words, the value that you set is the minimum number of bytes that a single packet must be before the server compresses it.

Many of the XML messages being sent around the system are greater than 100,000 bytes. These messages contain HTML in compressed states, so it's generally not required that these messages be compressed. Because of this, the PeopleSoft default is set to 1,000,000 bytes.

Be careful when adjusting compression settings. If you set the threshold too high, then no packets will be large enough to be compressed. If you set the threshold too low, you may greatly reduce network traffic, but be aware that the server will have an increased workload from compressing numerous packets. Typically, you should decrease the threshold according to the bandwidth of the workstation hardware as described in the following paragraphs.

If you are handling only LAN connections, you may want to disable compression by setting the threshold to 99999999 so that only packets larger than 99,999,999 bytes are compressed. Of course, such a large value effectively disables compression so that no packets are compressed. This means no extra work for the server compressing packets.

Alternatively, if you have mostly low bandwidth, as in 56-kilobyte (KB) modem connections over a WAN, then you most likely want to compress the packets as much as possible. When decreasing the compression threshold, keep in mind that the law of diminishing returns applies. Setting the threshold much below 1000 puts an increasing load on the server, and this can nullify any performance increases that you may have gained from reduced network traffic.

Click to jump to top of pageClick to jump to parent topicAdditional Prompt

After you finish all of the configuration sections, PSADMIN prompts you to configure Jolt (which is enabled by default).

If you are using the PeopleSoft Internet Architecture, you must have Jolt enabled for browser access.

Click to jump to parent topicJolt Relay Adapter Options

The Jolt relay adapter is disabled by default. Unless you have a specific need for JRAD, you should skip this section.

See Also

Understanding Jolt Internet Relay

Click to jump to top of pageClick to jump to parent topicListener Address

The default is %PS_MACH%. Enter the machine on which the application server is running. See the equivalent parameter for the workstation listener.

Click to jump to top of pageClick to jump to parent topicListener Port

This option is for advanced configurations requiring the Jolt internet relay (JRLY). The listener port listens for JRLY requests and must match the JRLY “OUT” port setting in the JRLY configuration file of the sending machine. The port number, as in 9100, is not used unless you enter y at the prompt that asks if you want to configure JRAD.

Click to jump to parent topicDomain Settings

Use this section to specify general settings for the entire domain—not just for a specific component of the domain.

Click to jump to top of pageClick to jump to parent topicDomain ID

Enter the name of the application server domain. It does not need to match the name that you specified when you created the domain. This name is important only because the Tuxedo Web Monitor and PeopleSoft Watch Server (PSWATCHSRV) use it to identify application server domains and the processes associated with each machine. It should not exceed 8 characters. Generally, you should use the database name in lowercase.

Click to jump to top of pageClick to jump to parent topicAdd to PATH

Enter the directory that contains your database connectivity software, as in /apps/db/oracle/bin, in the path. If the database connectivity directory is not already specified in the path, you can set it by specifying this parameter. The value is added to the path.

On Microsoft Windows, if you don't enter a value, it uses the current path.

On UNIX, if you don't enter a value, it uses the current directory—not the current path. To have it set by default to the current path, enter a period (.).

Note. On Windows, entries that contain a space must be surrounded by quotes.

Click to jump to top of pageClick to jump to parent topicSpawn Threshold

Enter a parameter that's supplied to Tuxedo for control of process spawning by using the -p command-line option for all server processes. The default setting (1,600:1,1) rarely needs to be changed.

This setting enables the dynamic decay of spawned server processes as the transaction volume decreases. The value can be loosely translated to mean that if, in 600 seconds, there is less than or equal to one job in the queue, the decay process begins. This is described in more detail in the timeout settings appendix of this PeopleBook.

New server processes will be spawned according to the rule defined here. By default, if there is one outstanding request in the queue for one second or more, an additional process is spawned. Additional processes will be spawned all the way up to the Max Instances defined for that server type. If Max Instances and Min Instances are identical, this setting has no effect.

Note. This parameter applies only if, for PSAPPSRV, the value of Max Instances is greater than that of Min Instances. By default, spawning is disabled.

See Also

Application Server Timeouts

For more information, see servopts(s) in the Tuxedo reference manual

Click to jump to top of pageClick to jump to parent topicLog Directory

The log directory contains log files the system generates for a domain, such as Tuxedo logs (TUXLOG) and APPSRV logs.

The default log directory for a domain is %PS_SERVDIR\logs.

To specify a custom log directory:

  1. Uncomment the Log Directory setting in the domain's PSAPPSRV.CFG file.

  2. Enter the desired log directory location either directly into the PSAPPSRV.CFG file or through PSADMIN.

  3. Reconfigure the domain using PSADMIN for the new setting to take effect.

When entering custom log directory locations, keep the following length restrictions in mind.

Domain Type

Log Directory Character Length Limit

Application Server

60

Process Scheduler Server

70

Search Server

80

Click to jump to top of pageClick to jump to parent topicRestartable

Enter y to have Tuxedo restart server processes (except the BBL process) if the server dies abnormally, as in a 'kill' on UNIX or through the Task Manager on Microsoft Windows. Otherwise, enter n.

Click to jump to top of pageClick to jump to parent topicAllow Dynamic Changes

Often, administrators must set a trace or performance parameter while the domain is up and running. If you enable this option, then you don't need to reboot the domain for the modified parameter value to take effect.

Enter y or n to enable or disable dynamic changes. When disabled, you must reboot (or cycle the processes) for changes to take effect.

When enabled, the server checks an internal time stamp for a particular service request to see if any values have changed for the parameters for which dynamic changes are valid. If values have changed, the system uses the modified parameter value.

You should enable this option in your test and development domains. For production environments, you should enable dynamic changes selectively.

These parameters allow dynamic changes:

Note. The parameters that allow dynamic changes are also identified through comments in the PSAPPSRV.CFG file. Look for the phrase “Dynamic changes allowed for X,” where X is the parameter name. This option does not apply to configuration parameters that Tuxedo relies on, such as the number of processes, whether restart is enabled, the port numbers, the amount of handlers, and so on.

Click to jump to top of pageClick to jump to parent topicLogFence

Enter a level of network tracing, ranging from –100 (suppressing) to 5 (all). The default is 3.

The trace file is generated in PS_CFG_HOME\appserv\domain\LOGS\psappsrv.log.

Click to jump to top of pageClick to jump to parent topicAppLogFence

This setting is not available through the PSADMIN interface, but can be entered directly into the PSAPPSRV.CFG file.

You can use this parameter conditionally to determine whether you want to do specific logging from your application. You can implement this parameter from PeopleCode using the %AppLogFence system variable. Complete documentation for this option is in the PeopleCode Developer's Guide.

See Using Application Logging.

Click to jump to top of pageClick to jump to parent topicTrace-Log File Character Set

Enter the character set (ANSI or UNICODE) of the machine to which you typically write and read the traces and log files. If the character sets are not matched between the file and the machine, the file is unreadable.

Click to jump to parent topicPeopleCode Debugger Options

Use this section to enable and configure the PeopleCode debugging environment. Configuring PeopleCode debugging is discussed in detail in another section of this PeopleBook.

See Also

Setting Up the PeopleCode Debugger

Click to jump to parent topicTrace Options

This section enables you to specify the tracing options that you can enable on the application server to track the Structured Query Language (SQL) and PeopleCode of the domains. You can also set all of the trace parameters from the PeopleSoft sign-in page. Just beneath the Sign In button, click the link that opens the trace flags page. This enables you to set the trace options and then sign in to the system.

Note. With many of the following trace options, you need to view the comments in the PSAPPSRV.CFG to understand what to enter to return the trace information you require.

Click to jump to top of pageClick to jump to parent topicTraceSQL

Enter the logging level for SQL tracing for all clients. Traces are written to PS_CFG_HOME/appserv/domain/LOGS/domain_user_IDservername.tracesql.

Enter 0 to disable tracing; enter 7 to enable a modest tracing level for debugging. For other levels of tracing, set this option to a value that equals the sum of the needed options. For example, to trace only SQL, enter 1; to trace SQL statements and connect statements enter 7 (1+ 2 + 4 = 7). A setting of 7 is recommended for troubleshooting connection and other basic problems. Tracing can consume large amounts of disk space over time, so be sure to reset this option to 0 when you finish troubleshooting.

Important! The trace file stores elapsed times for SQL events to a precision of one microsecond (six decimal places). However, due to limitations of the operating system, Windows precision is actually in milliseconds (three decimal places), so the last three digits in a Windows trace will always be zero. Elapsed times in UNIX are accurate to one microsecond.

Click to jump to top of pageClick to jump to parent topicTraceSQLMask

Enter the logging level ceiling for SQL tracing for individual clients. Traces are written to PS_CFG_HOME/appserv/domain/LOGS/client_user_IDservername.tracesql.

For Windows clients, you specify the necessary SQL tracing level by using the PeopleSoft Configuration Manager on the Trace tab. To prevent clients from turning on the application server trace and consuming resources, the application server uses TraceSQLMask as an administrative control facility.

If a client transmits a request to trace SQL, the application server compares the value that is transmitted to the TraceSQLMask value. If the client value is less than or equal to the TraceSQLMask value, the application server enables the trace. However, if the client value is greater, the application server enables the trace up to the TraceSQLMask value. Trace files are written on the application server; no trace shows up on the client workstation.

Click to jump to top of pageClick to jump to parent topicTracePC

Enter a level for PeopleCode tracing for activity that is generated by all clients on a domain. Eligible values are defined in the configuration file. TracePC values are displayed in the PeopleSoft Configuration Manager on the Trace tab. You can find the results in PS_CFG_HOME/appserv/domain/LOGS/domain.log.

Important! The trace file stores elapsed times for PeopleCode events to a precision of one microsecond (six decimal places). However, due to limitations of the operating system, Windows precision is actually in milliseconds (three decimal places), so the last three digits in a Windows trace will always be zero. Elapsed times in UNIX are accurate to one microsecond.

Click to jump to top of pageClick to jump to parent topicTracePCMask

Enter which PeopleCode trace options that are requested by client machines will be written to the trace file. You can find the results in PS_CFG_HOME/appserv/domain/LOGS/client_machine.domain.log.

Click to jump to top of pageClick to jump to parent topicTracePPR and TracePPRMask

Use these options to trace the activity in the page processor. Typically, these options are used internally only by PeopleSoft developers; however, you may need to view the results of this trace when troubleshooting.

Tracing-related display processing is useful for seeing when and if related displays are being updated and if they are updating successfully. Some sample output in the log file from setting this flag includes:

Starting Related Display processing Related Display processing - PPR_RELDSPLVALID not set Related Display processing - All Rows Starting Related Display processing for - PSACLMENU_VW2.MENUNAME Related Display processing for - PSACLMENU_VW2.MENUNAME - completed successfully Finished Related Display processing

By using the keylist generation tracing in addition to the related display tracing, you can determine why the related displays have the wrong value. It shows where the keys are coming from. The following is a sample of keylist generation tracing:

Starting Keylist generation Keylist generation - FIELDVALUE is a key FIELDVALUE is low key Low key value was supplied = Key FIELDVALUE = Keylist generation - FIELDNAME is a key Keylist generation - Finding value for USRXLATTABLE_VW.FIELDNAME Not Found in key buffer Seaching for field FIELDNAME in component buffers Scanning level 1 Scanning record DERIVED_USROPTN for field FIELDNAME Field FIELDNAME found in record DERIVED_USROPTN Found in component buffers, value = PT_TIME_FORMAT Key FIELDNAME = PT_TIME_FORMAT Keylist generation - USEROPTN is a key Keylist generation - Finding value for USRXLATTABLE_VW.USEROPTN Not Found in key buffer Seaching for field USEROPTN in component buffers Scanning level 1 Scanning record DERIVED_USROPTN for field USEROPTN Scanning record PSUSROPTLIST_VW for field USEROPTN Field USEROPTN found in record PSUSROPTLIST_VW Found in component buffers, value = TFRMT Key USEROPTN = TFRMT Keylist Generation complete FIELDNAME = PT_TIME_FORMAT FIELDVALUE = USEROPTN = TFRMT

In this example, you can see how the system builds the keylist by first searching in the current record (key buffer), then searching the buffers in the current level, and then searching up a level, and so on. It also indicates exactly what record the key value is taken from. This is useful on complex components where there are often several instances of a particular field; a common problem is that the value is derived from an unexpected location.

Combining the keylist tracing and the related display tracing provides a good view of the system behavior. For example:

Starting Related Display processing Related Display processing - All Rows Starting Related Display processing for - PSACLMENU_VW2.MENUNAME Starting Keylist generation Keylist generation - MENUNAME is a key MENUNAME is low key Low key value was supplied = APPLICATION_ENGINE Key MENUNAME = APPLICATION_ENGINE Keylist Generation complete MENUNAME = APPLICATION_ENGINE Related Display processing for - PSACLMENU_VW2.MENUNAME - completed successfully

Each related display goes through the keylist generation process, and you can see exactly what key values are used to populate the related displays and where those key values came from.

The work record flag is a performance feature. If every field in a level-0 record has a value from the keylist and is display-only, then it is marked as a work record because the values can't be changed. After it is marked as a work record, that affects how the record behaves. For example, PeopleCode for fields in the record but not in the component don't run, data isn't saved, and so on. By enabling this tracing option, you can see which records are flagged as work records. The output looks like this:

Work flag cleared for record PSCLASSDEFN_SRC Work flag cleared for record PSCLASSDEFN_SRC Work flag cleared for record PSCLASSDEFN Work flag cleared for record PSPRCSPRFL Work flag cleared for record SCRTY_QUERY Work flag set for record PSCLASSDEFN Work flag set for record PSPRCSPRFL Work flag set for record SCRTY_QUERY

Because the flag is turned on and off at various points, the last setting (set or cleared) is the most important. In the previous trace, PSCLASSDEFN, which is marked as a work record, is cleared and then set again.

Click to jump to top of pageClick to jump to parent topicTracePIA and TracePIAMask

Use these options for tracing PeopleSoft page (PIA page) generation.

Click to jump to top of pageClick to jump to parent topicTraceAE

Use this parameter to activate specific Application Engine traces for tracing Application Engine programs.

Click to jump to top of pageClick to jump to parent topicTraceAnalytic and Trace AnalyticMask

The bits enable logging for analytic servers beyond the standard LogFence setting.

Click to jump to top of pageClick to jump to parent topicTracePPM

The Performance Monitor agent is a thread that reports performance metrics for each instrumented server if monitoring is enabled for the database. Select 1 to enable and 0 to disable.

See Working with the Performance Trace.

Click to jump to top of pageClick to jump to parent topicDumpMemoryImageAtCrash

This parameter determines whether or not a memory image of the failing process is created when a crash occurs. By default, the value is 'NONE' which means that a memory image will not be written during a crash. This value can be set to 'MINI' or 'FULL'. MINI means that a shorter memory image is written. This may be a better option if you are leaving this option turned on permanently. FULL may be appropriate when you are debugging a known issue. Typically, this option is used internally only by PeopleTools.

Click to jump to top of pageClick to jump to parent topicDumpManagerObjectsAtCrash

This parameter assists PeopleSoft in debugging any crash problems at your site by providing insight into the customized object definitions to reproduce the crash on another database.

Click to jump to top of pageClick to jump to parent topicLog Error Report, Mail Error Report

If you enter y (enabled) and runtime errors are detected (nonfatal error conditions), the system writes a message and information regarding the runtime error to the current log file. If you assign the MailErrorReport parameter an email address, an individual, such as a system administrator, can be alerted whenever the system writes an error to the log. If MailErrorReport is enabled but LogErrorReport is set to n, the system still sends a message for application server crashes, which always cause data to be written to the log. The following is an example of setting this parameter to send notifications to an email address: MailErrorReport=tom.sawyer@bigcompany.com.

Click to jump to top of pageClick to jump to parent topicWrite Crash Dump to Separate File

If the application server shuts down abnormally, you can view the log information that is related to the shutdown. However, because this information can be lengthy, this option enables you to write the information to a file other than the appserv.log file. To enable this option, enter y.

The system writes the crash dump file to PS_CFG_HOME\appserv\domain\logs. The system names the crash dump file according to the following convention: server_process_name.process_ID.dmp.

The following example shows what appears in the appserv.log in the event of a crash:

(0) Unhandled exception occurred. Writing crash dump to PSAPPSRV.213.dmp (3) Switching to new log file b:\ptserver\appserv\test\logs\PSAPPSRV.213.dmp

To disable this option, enter n. If you do not enable this option, crash information appears in the appserv.log by default.

Click to jump to parent topicCache Settings

Use this section to specify how to handle caching at your site. Enabling caching on the application server improves performance, but you need to review the available cache options and determine which option best suits your site.

Note. The majority of the cache settings need to be uncommented in the PSAPPSRV.CFG file.

Click to jump to top of pageClick to jump to parent topicEnableServerCaching

With EnableServerCaching, you specify what objects the system stores in cache on the application server. To enable application server disk caching the value must be set to 1 or 2.

If you enter 1 the system caches only the most used classes of objects, and if you enter 2, the system caches all object types regardless of the frequency of use. Which option you select depends on internal testing at your site.

To disable application server caching, set this value to 0. In most cases there is no reason to disable server caching. Doing so significantly degrades performance, because it requires the application server to retrieve an object from the database each time the system needs it.

Click to jump to top of pageClick to jump to parent topicCacheBaseDir

This setting is the location where cache files will be written and stored for this domain.

Note. You must preload your shared cache before you enable shared caching on the application server.

Application Engine processes are independent from application server domains, directories, and configuration files. Therefore, Application Engine processes do not share cache with application server domain processes.

Click to jump to top of pageClick to jump to parent topicServerCacheMode

Caching enables the system to store definitions locally on the application server, eliminating unnecessary trips to the database server. If caching were disabled, the system would need to retrieve definitions from the database with each request, every time. This adds significant overhead to each transaction and affects system response times.

If server caching is enabled on the application server, which is usually the recommended approach, there are two modes of caching from which to choose.

Cache Mode

Description

Non-shared cache

By default, non-shared cache mode is enabled (ServerCacheMode set to 0).

With the non-shared cache mode, each server process that starts within a domain maintains its own separate cache file.

For example, assume you had two PSAPPSRV processes configured in a domain. In non-shared cache mode, there is one cache directory per PSAPPSRV server process, which each individual PSAPPSRV process uses separately.

In this case, you can find the cache files in PS_CFG_HOME\appserv\domain\cache\PSAPPSRV_1 and \PSAPPSRV_2.

While the cache directories will grow over time to include the most used definitions, you have the option to preload the non-shared cache directories with the most used system definitions.

See Configuring an Application Server Domain to Preload Cache.

Shared cache

To set shared caching for the domain, enter 1 at the Set ServerCacheMode prompt. With this option enabled, you can find the cache files in PS_CFG_HOME\appserv\domain\cache\share.

The system assumes that a preloaded cache exists in the share directory. The preloaded cache contains most instances of the managed object types that are cached to file. When you boot the application server, if shared cache files are enabled but no cache files exist in the expected location, the system reverts to unshared caching.

To preload shared cache, you run delivered Application Engine programs that build your shared cache.

See Load Application Server Cache.

Click to jump to top of pageClick to jump to parent topicMaxCacheMemory

PeopleTools stores metadata in a memory cache to increase system performance. However, too large a cache can leave insufficient available memory on your system, which leads to reduced performance.

Use this setting to specify the maximum size of the memory cache. Every time you use an object, its LastUsedDate value is updated. When your system reaches the memory cache threshhold, the system prunes the oldest objects in the cache first — that is, the ones with the oldest LastUsedDate values — and places the pruned data in a disk cache instead. It prunes the cache to keep it 10% below the specified threshhold.

Because using a disk cache can also reduce performance, the default setting might not be optimal for your application. You can adjust this setting to achieve the best trade-off between speed and available memory.

Enter an integer value to specify the maximum size of the memory cache in megabytes. By specifying a value of 0 megabytes you disable pruning altogether, which allows for an unlimited memory cache. The default value of this setting is 10 megabytes.

Click to jump to top of pageClick to jump to parent topicEnableDBCache

If enabled, the metadata (cache) is accessed from database, rather than the file system.

To implement database caching, uncomment the parameter and enter Y to enable database caching, or N to disable database caching. After changing this parameter, the domain does not need to be reconfigured.

The database cache is shared by all domains that enable this option.

When database caching is enabled, these settings are ignored:

You can load the database cache using the Load Application Server Cache utility or the preload cache utility.

Note. Database caching is also available for Process Scheduler domains.

See Also

Load Application Server Cache

Configuring an Application Server Domain to Preload Cache

Using the PSADMIN Utility

Click to jump to top of pageClick to jump to parent topicPreLoadCache and PreLoadMemoryCache

If you have created a cache project, specify the project name. Before booting a domain, you have the option to preload the cache, and this option creates the cache based on the load project you specify.

If database caching is enabled, the cache is preloaded into the database.

See Also

Configuring an Application Server Domain to Preload Cache

Click to jump to parent topicRemote Call Options

There are two significant Remote Call domain parameters.

Click to jump to top of pageClick to jump to parent topicRCCBL Redirect

You must set the RCCBL Redirect option for remote call through PSADMIN.

Enter 0 to disable redirection and 1 to enable redirection. Redirection causes the server process to retain intermediate work files that are used to pass parameter values between the server process and a RemoteCall/COBOL program for debugging purposes.

Note. Redirect should always be enabled because it contains valuable information for debugging purposes.

The location of the intermediate Remote Call files is $PS_CFG_HOME/appserv/domain/log_output. The intermediate Remote Call files generated are:

Note. Where <pid> is the Process ID.

Click to jump to top of pageClick to jump to parent topicRCCBL PRDBIN

Use this parameter to specify where RemoteCall can find the COBOL executables. By default, RemoteCall looks in a predefined location. This might not be the location where you've installed them on your system:

To override this default behavior, set RCCBL PRDBIN to the absolute path of your COBOL executables, for example:

Note. This parameter doesn't appear in the PSADMIN custom configuration interface if it's not already set. You must define it by editing the application server configuration file directly. On the PeopleSoft Domain Administration menu, select Edit configuration/log files menu, then select Edit psappsrv.cfg (current configuration file) to open psappsrv.cfg in a text editor. Define the RCCBL PRDBIN parameter in the RemoteCall section of the file.

Click to jump to parent topicPSAPPSRV Options

The PSAPPSRV server process performs the functional requests, such as building and loading panel groups. It also provides the in-memory-caching feature for PeopleTools objects on the application server. Each server process maintains its own cache.

Click to jump to top of pageClick to jump to parent topicMin Instances

Enter the minimum number of application server instances that start when you boot the domain. There's always at least this number of instances running. This translates to the PSAPPSRV server's -m (min) parameter in the UBB file.

Click to jump to top of pageClick to jump to parent topicMax Instances

Enter the maximum number of server instances that can be started. This translates to the PSAPPSRV server's -M (Max) parameter in the UBB file.

Click to jump to top of pageClick to jump to parent topicService Timeout

Enter the number of seconds that a PSAPPSRV waits for a service request, such as MgrGetObj or PprLoad, to complete before timing out. Service timeouts are recorded in the TUXLOG and APPSRV.LOG. In the event of a timeout, PSAPPSRV is terminated and a replacement process is started by Tuxedo.

Click to jump to top of pageClick to jump to parent topicRecycle Count

Enter the number of service requests that each server has carried out before being terminated (intentionally) and then immediately restarting. Servers must be intermittently recycled to clear buffer areas. The time that is required to recycle a server is negligible, occurring in milliseconds. The recycle count does not translate into a native Tuxedo parameter in the PSAPPSRV.UBB file. Instead, the value is stored in memory and is managed by a PeopleSoft server.

Click to jump to top of pageClick to jump to parent topicPercentage of Memory Growth

The Percentage of Memory Growth option, enables you to perform dynamic recycling, in a test environment, so that you can arrive at a static Recycle Count value suited to your production system. The Percentage of Memory Growth parameter indicates the percentage of memory growth to reach before the PSAPPSRV process will automatically restart (dynamically recycle). The default is 20, meaning an additional 20% of memory growth will be incurred after the process has established a baseline memory cache.

A dynamic recycling configuration is intended to be used in testing environment where a usage load, representative of your production usage load, can be run against a system. During this load test, administrators can monitor log files to determine when the dynamic recycle occurs. The intervals and trends related to the dynamic recycling can help to identify an appropriate Recycle Count value for a production environment. When the load test is complete, the Recycle Count value should be adjusted appropriately, and the dynamic recycle feature should be disabled.

When dynamic recycling is configured, the system evaluates the memory size after every 100 service requests to determine if the process needs to be recycled. If any new PeopleTools objects have been loaded into the memory cache, or if a JVM has been initialized since the last memory evaluation, the memory baseline is reset to the current value and no recycling will occur.

If no PeopleTools objects have been loaded, and if the JVM has not been initialized since the last memory evaluation, and if the memory footprint has grown by more than the specified percentage, the system recycles the PSAPPSRV process and adds messages to the APPSRV_xxyy.log file.

Note. Due to the overhead involved in measuring the memory usage, dynamic recycling is not recommended for use in a production environment.

Note. Unless you can emulate, in your test environment, a usage load representative of a typical production usage load, the results of your test will be of little value for determining the optimal Recycle Count value.

Configuring Dynamic Recycling

To configure dynamic recycling:

  1. Open PSAPPSRV.CFG.

  2. In the PSAPPSRV section, set Recycle Count to 0 to disable the fixed recycle interval.

  3. Uncomment the Percentage of Memory Growth parameter, and adjust the value as needed to a value between 1–100, where 100 = 100%.

    Percentage of Memory Growth=20

  4. Restart the domain.

Viewing Logs for Dynamic Recycle Count Information

Recycle messages are logged with the service count indicated at that point. The system logs messages based on the following scenarios:

Memory Evaluation Scenario

Message Logged

Memory could not be determined.

Use Recycle Count for recycling due to unable to obtain virtual memory size

In this case, the message is logged at the "Status Level". This message appears at LogFence=3.

The current service count is less than the Recycle Count, although the memory usage exceeds the Percentage of Memory Growth value.

This indicates that the Recycle Count was set to a value other than 0. Should this condition occur, the system logs this message:

delay dynamic recycle(services=100, Recycle Count=500)

At intervals of 100 and if the LogFence is set to at least 4.

dynamic recycle results [ recycle=true, orig_mem=1000000, current_mem=2000000, max_mem=3000000, check_count=100, obj_loaded=100, jvm_loaded=true ]

Click to jump to top of pageClick to jump to parent topicAllowed Consec Service Failures

Enter a number greater than 0 to enable dynamic server processes to restart for service failures. To disable this option, enter 0. The default is 2. The value that you enter is the number of consecutive service failures that will cause a recycle of the server process. This is a catchall error handling routine that enables PSAPPSRV, PSQCKSRV, and PSAMSRV to terminate themselves if they receive multiple, consecutive, fatal error messages from service routines. Such errors should not occur consecutively, but if they do, the server process must be recycled or cleansed. A retry message appears on the client browser when this occurs.

Click to jump to top of pageClick to jump to parent topicMax Fetch Size

The default is 5000 (K). Enter the maximum memory that is used by the server to store fetched rows for a transaction before sending the result set back to a client. If the memory limit is exceeded, the client receives the rows retrieved with a memory buffer exceeded warning. You should use the default value. PSAPPSRV supports nonconversational transactions, so this parameter provides a way to balance high-volume throughput with the needs of users working with large volumes of data. A value of 0 means unlimited memory is used. The memory is not preallocated, but it is acquired as needed for each transaction.

Click to jump to top of pageClick to jump to parent topicAuto Select Prompts

Enter 1 (the default) to enable automatic prompting on lookup pages. When the user selects the prompt lookup button, the application server automatically returns all values for that field, up to 300 rows. If necessary, the user can refine the search further by entering partial data in the Search By field.

Enter 0 to require the user to enter a partial value before the automatic prompt list appears.

Click to jump to top of pageClick to jump to parent topicAutoLoad JVM

You can add this parameter, if needed, manually to the [PSAPPSRV] section of the PSAPPSRV.CFG file. AutoLoad JVM controls whether the JVM gets loaded automatically when the domain boots.

By default, domain behavior reflects a setting of AutoLoad JVM=0 (not enabled).

If you have configured EnablePPM Agent=1, then JVM will be loaded at the domain boot time. If you have configured EnablePPM Agent=0, then to load JVM when the domain boots, you need add AutoLoad JVM=1 to the [PSAPPSRV] section of the PSAPPSRV.CFG file.

Click to jump to top of pageClick to jump to parent topicSerial Recycle

When serial recycling is enabled for a server process, the system recycles server processes of that type within a domain on a serial basis–one after another–to allow processing to continue uninterrupted. By default, the domain behavior reflects Serial Recycle=Y (enabled).

When Serial Recycle=Y for PSAPPSRV, for example, then only one PSAPPSRV process will recycle during the recycle time of 60 seconds when the Recycle Count limit is reached. Then the next PSAPPSRV will recycle.

When serial recycling is not enabled, all the server processes of that type recycle simultaneously when the Recycle Count limit is reached, which can cause throughput to pause.

Serial recycling applies to these server process types: PSAPPSRV, PSQCKSRV, PSQRYSRV, and the Integration Broker server processes.

To disable serial recycling, manually add the Serial Recycle parameter, and assign the value N. For example:

Serial Recycle=N

The recycle time is 60 seconds.

Click to jump to parent topicPSANALYTICSRV Options

PSANALYTICSRV relates to the server processes that are associated with the analytic server framework. Most of the parameters for PSANALYTICSRV need to be reviewed in the PSAPPSRV.CFG file and uncommented as needed.

See Also

Managing Analytic Servers

Click to jump to top of pageClick to jump to parent topicMin Instances

Enter the minimum number of analytic server instances that start when you boot the application server domain. There's always at least this number of instances running. The default value of this parameter is 3.

Click to jump to top of pageClick to jump to parent topicMax Instances

Enter the maximum number of analytic server instances that can result from dynamically spawning new processes. The default value of this parameter is 3.

See Managing Analytic Servers.

Click to jump to top of pageClick to jump to parent topicAnalytic Instance Idle Timeout

Enter the number of minutes of inactivity before the analytic instance times out and is unloaded.

This value takes effect only if the PeopleCode AnalyticInstance class Load method specifies a value of -1 for its IdleTimeOut parameter when loading an analytic instance. This includes Load PeopleCode that's launched from an analytic grid, which enables you to avoid having to explicitly specify a timeout.

The default value of this parameter is 0 (no timeout limit) for domains that are configured with a developer template, and 30 minutes for other domains.

Click to jump to parent topicPSSAMSRV Options

The PSSAMSRV server process communicates through the Tuxedo conversational mode. It performs transactional SQL requests (updates).

Click to jump to top of pageClick to jump to parent topicMin Instances

Enter how many servers are started at boot time. This translates to the PSSAMSRV server's -m (min) parameter in the UBB file.

Click to jump to top of pageClick to jump to parent topicMax Instances

Enter the maximum number of servers that can be started. This translates to the PSSAMSRV server's -M (Max) parameter in the UBB file.

Click to jump to top of pageClick to jump to parent topicService Timeout

Enter the number of seconds that the server processes waits for a request before timing out. This stops runaway processes, like an rccbl timeout.

Click to jump to top of pageClick to jump to parent topicRecycle Count

Enter the number of service requests that each server carries out before being terminated (intentionally). Tuxedo immediately restarts the server. Servers must be intermittently recycled to clear buffer areas. The time that is required to recycle a server is negligible, occurring in milliseconds. The recycle count does not translate into a native Tuxedo parameter in the PSAPPSRV.UBB file. Instead, the value is stored in memory and is managed by a PeopleSoft server.

Click to jump to top of pageClick to jump to parent topicAllowed Consec Service Failures

Enter a number greater than zero to enable dynamic server process restarts for service failures. To disable this option, enter 0. The default is 2. The value that you enter is the number of consecutive service failures that cause a recycle of the server process. This is a catchall error handling routine that enables PSAPPSRV, PSQCKSRV, and PSSAMSRV to terminate themselves if they receive multiple, consecutive, fatal error messages from service routines. Such errors should not occur consecutively, but if they do, the server process must be recycled or cleansed. A retry message appears on the client browser when this occurs.

Click to jump to top of pageClick to jump to parent topicMax Fetch Size

The default is 32 (K). Enter the maximum memory that is used by the server to store fetched rows for a transaction before sending results to the client and refilling the memory buffer. When the memory limit is reached, the server sends rows to the client, but then resumes refilling the buffer and sending results to the client until the query is complete. You should leave the default value unchanged.

PSSAMSRV supports conversational transactions, so this parameter enables users to tune performance by adjusting the number of network round-trips that are required for the average transaction. A value of 0 causes unlimited memory to be used, which means one round-trip no matter how large the result set. The memory is not preallocated, but is acquired as needed.

Click to jump to parent topicPSQCKSRV Options

The PSQCKSRV is an optional server process to improve performance. Essentially, the PSQCKSRV, or quick server, is a copy of the PSAPPSRV. It performs quick requests, such as nontransactional (read-only) SQL requests. The PSQCKSRV improves overall performance by enabling the PSAPPSRV process to direct a portion of its workload to PSQCKSRV.

Click to jump to top of pageClick to jump to parent topicMin Instances

Enter how many servers are started at boot time. This translates to the PSQCKSRV server's –m (min) parameter in the UBB file.

Click to jump to top of pageClick to jump to parent topicMax Instances

Enter the maximum number of servers that can be started. This translates to the PSQCKSRV server's –M (Max) parameter in the UBB file.

Click to jump to top of pageClick to jump to parent topicService Timeout

Enter the number of seconds that a PSQCKSRV waits for a request before timing out. This stops runaway processes, like an rccbl timeout. This applies to incremental PSQCKSRV servers that are dynamically started by the Max Instances parameter.

Click to jump to top of pageClick to jump to parent topicRecycle Count

Use the PSAPPSRV specifications.

Click to jump to top of pageClick to jump to parent topicAllowed Consec Service Failures

Enter a number greater than zero to enable dynamic server process restarts for service failures. To disable this option, enter 0. The default is 2. The value that you enter is the number of consecutive service failures that will cause a recycle of the server process. This is a catchall error handling routine that enables PSAPPSRV, PSQCKSRV, and PSAMSRV to terminate themselves if they receive multiple, consecutive, fatal error messages from service routines. Such errors should not occur consecutively, but if they do, the server process must be recycled or cleansed. A retry message appears on the client browser when this occurs.

Click to jump to top of pageClick to jump to parent topicMax Fetch Size

Use the PSAPPSRV specifications.

Click to jump to top of pageClick to jump to parent topicSerial Recycle

Use the PSAPPSRV specifications.

Click to jump to parent topicPSQRYSRV Options

PSQRYSRV handles the SQL that is generated by PeopleSoft Query (PSQED.EXE). With PSQRYSRV configured, SQL-intensive, complicated, user-defined queries are offloaded to a dedicated server process, thus freeing PSAPPSRV and PSQCKSRV to handle the SQL requests for which they are more suited.

Note. When running PS/nVision reports from a three-tier, Windows client connection, the system also routes the SQL generated by both matrix (ledger-based) and tabular (PS Query-based) reports through PSQRYSRV if it is enabled.

PSQCKSRV also processes SQLRequest services; however, if PSQRYSRV is configured, it processes all SQLRequests that are initiated specifically by PSQuery (SQLQuery:SQLRequest) or PS/nVision.

Like the PSQCKSRV server process, PSQRYSRV is an optional server process. However, if you allow users to initiate queries from PeopleSoft Query, you should take advantage of this server process.

Click to jump to top of pageClick to jump to parent topicMin Instances

Enter how many servers are started at boot time. This translates to the PSQRYSRV server's –m (min) parameter in the UBB file.

Click to jump to top of pageClick to jump to parent topicMax Instances

Enter the maximum number of servers that can be started. This translates to the PSQRYSRV server's –M (Max) parameter in the UBB file.

Click to jump to top of pageClick to jump to parent topicService Timeout

Enter the number of seconds that PSQRYSRV waits for a request before timing out. This stops runaway processes.

Click to jump to top of pageClick to jump to parent topicRecycle Count

Enter the number of service requests that each server carries out before being terminated (intentionally) by Tuxedo and then immediately restarted. Servers must be intermittently recycled to clear buffer areas. The time that is required to recycle a server is negligible, occurring in milliseconds.

If the recycle count is set to 0, PSQRYSRV is never recycled.

Click to jump to top of pageClick to jump to parent topicAllowed Consec Service Failures

Enter a number greater than 0 to enable dynamic server process restarts for service failures. To disable this option, enter 0. The default is 2. The value that you enter is the number of consecutive service failures that will cause a recycle of the server process. This is a catchall error handling routine that enables PSAPPSRV, PSQCKSRV, PSQRYSRV, and PSSAMSRV to terminate themselves if they receive multiple, consecutive, fatal error messages from service routines. Such errors should not occur consecutively, but if they do, the server process must be recycled or cleansed. A retry message appears on the client browser when this occurs.

Click to jump to top of pageClick to jump to parent topicMax Fetch Size

Enter the maximum size (in KB) of a result set that is returned from a SELECT query. The default is 10000 KB. Use 0 for no limit.

Click to jump to top of pageClick to jump to parent topicUse Dirty-Read

(Applies only to DB2 systems.) Enter 1 to enable the application server to read uncommitted data from a table. Enter 0 to disable dirty reads.

This parameter can be used for general reporting and PeopleSoft Query. You can run dirty read queries through the application server, the Process Scheduler, and in a two-tier connection. The Use Dirty-Read setting in the application server configuration controls the behavior of PSAPPSRV, PSQCKSRV, and PSQRYSRV.

Note. Dirty reads are not recommended if you are reading data and doing subsequent processing based on the disposition of the data at the time that it is read. Between the time the data is read by a subsequent process and the time the unit of work is completed by the first process, any activity affecting the table data at the time a subsequent process read could be rolled back, invalidating the accuracy of the data that a subsequent process read.

See Also

Creating and Running Simple Queries

Setting Parameters for the Application Engine Server

Click to jump to top of pageClick to jump to parent topicSerial Recycle

Use the PSAPPSRV specifications.

Click to jump to parent topicIntegration Broker Server Processes

A variety of server processes are devoted to Integration Broker processing. These servers are configured in the Publish&Subscribe PSMSGDSP, and PSMSGHND sections. If you are not implementing the Integration Broker technology, skip through these delivered, default server processes:

These server processes act as dispatchers and handlers of the messages in the messaging system.

See Also

Administering Messaging Servers for Asynchronous Messaging

Click to jump to parent topicSMTP Settings

You can send electronic mail requests, issued with workflow or PeopleCode, to the application server, which passes the requests to the specified mail server (SMTPServer). To specify the appropriate SMTP server and port to receive the email requests, you must edit the SMTP Settings section.

When set in the PSAPPSRV.CFG file, these SMTP settings are not dynamic: SMTPGuaranteed, SMTPTrace, SMTPSendTime. They require a domain reboot to take effect.

Note. You can also control most of these settings using the PeopleCode SMTPSession class, which temporarily overrides them without changing the values in PSAPPSRV.CFG. You use this class primarily when you want to send multiple emails in a single session of the SMTP server, instead of having to change the permanent SMTP settings for every email. In the event that your application PeopleCode does not specify its own SMTP settings, the system uses the settings in the PSAPPSRV.CFG file.

See SMTPSession Class.

Click to jump to top of pageClick to jump to parent topicSMTPServer

Enter the host name and IP address of the mail server machine.

Click to jump to top of pageClick to jump to parent topicSMTPPort

Enter the port number on the mail server machine.

Click to jump to top of pageClick to jump to parent topicSMTPServer1

Enter the host name and IP address of the failover mail server machine in case the other specified server is down.

Click to jump to top of pageClick to jump to parent topicSMTPPort1

Enter the port number on the failover mail server machine.

Click to jump to top of pageClick to jump to parent topicSMTPSender

Enter the sender's internet address. This must be a valid address, such as user1@xyzcorp.com.

Click to jump to top of pageClick to jump to parent topicSMTP BlackberryReplyTo

Enter the internet address that you want to be the reply to address for Blackberry email responses. This must be a valid address such as user1@xyzcorp.com.

Click to jump to top of pageClick to jump to parent topicSMTPSourceMachine

Enter the sender's source machine name and internet address in the form of MACHINE.XYZCORP.COM. This value is required in some, but not all environments.

Click to jump to top of pageClick to jump to parent topicSMTPCharacterSet

Enter the character set that is used on the sender's machine.

Click to jump to top of pageClick to jump to parent topicSMTPEncodingDLL

Enter the name of a dynamic-link library (DLL) that is used to translate the mail message from the sender's character set (such as latin1, sjis, big5, gb, ks-c-5601-1987, or ks-c-5601-1992) to a 7-bit safe character set for transmission.

Click to jump to top of pageClick to jump to parent topicSMTPGuaranteed

Set this parameter to 1 if you want TriggerBusinessEvent email PeopleCode to be delivered through the Integration Broker system, which provides some additional administration capabilities for ensuring delivery of the message. By enabling this feature you implement a mechanism to ensure that emails get routed to the appropriate place in case SMTP mail fails for reasons such as network timeouts, downed mail servers, invalid parameters, and so on.

This setting enables the system to track email messages sent using Integration Broker queues. When sending an email with this option enabled, the system performs an asynchronous local-to-local publish, and for the subscription the system calls MCFOutboundMail.send to email the message.

Click to jump to top of pageClick to jump to parent topicSMTPTrace

Enter 1 to enable the tracing of email details to the log file. Enter 0 to disable SMTP tracing.

The system writes the information to the application server log file (APPSRV_MMDD.LOG) in %PS_SERVDIR%/LOGS, by default, or the custom value set for Log Directory.

LogFence should be set to 5.

Click to jump to top of pageClick to jump to parent topicSMTPSendTime

Enter 1 to have messages contain a send time that is populated by the application server. Enter 0 to leave the send time blank and have it populated by the receiving gateway (depending on the gateway).

Click to jump to top of pageClick to jump to parent topicSMTPUserName

Enter the user name to log in to the SMTP server. This applies only when authentication is enabled on the SMTP server.

Click to jump to top of pageClick to jump to parent topicSMTPUserPassword

Enter the password for the user specified by SMTPUserName to access the SMTP server. This applies only when authentication is enabled on the SMTP server.

Click to jump to top of pageClick to jump to parent topicSMTPUserName1

Enter the user name to log in to the failover SMTP server. This applies only when authentication is enabled on the failover SMTP server.

Click to jump to top of pageClick to jump to parent topicSMTPUserPassword1

Enter the password for the user specified by SMTPUserName1 to access the failover SMTP server. This applies only when authentication is enabled on the failover SMTP server.

Click to jump to top of pageClick to jump to parent topicSMTPTimeToWaitForResult

Enter the time in milliseconds for the mail system to wait for the result of sending each email. If the time is set to 0, the system doesn't wait, and the returned result will be always be %ObEmail_SentButResultUnknown ( = -1). If you set this parameter to -1, the system will wait for the completion of the send process. The default value of this setting is 10000 (ten seconds).

Click to jump to top of pageClick to jump to parent topicSMTPUseSSL

Indicates that SSL connections are enabled. Enter Y for yes or N for no. The default value is N.

Click to jump to top of pageClick to jump to parent topicSMTPSSLPort

If using SSL, specify the SSL port on the SMTP server. The default is 465.

Click to jump to top of pageClick to jump to parent topicSMTPClientCertAlias

If the SMTP server is configured for client authentication, enter the alias name of the client certificate.

Click to jump to top of pageClick to jump to parent topicSMTPSSLPort1

Applies to the failover SMTP server.

If using SSL, specify the SSL port on the SMTP server. The default is 465.

Click to jump to top of pageClick to jump to parent topicSMTPUseSSL1

Applies to the failover SMTP server.

Indicates that SSL connections are enabled. Enter Y for yes or N for no. The default value is N.

Click to jump to top of pageClick to jump to parent topicSMTPClientCertAlias1

Applies to the failover SMTP server.

If the SMTP server is configured for client authentication, enter the alias name of the client certificate.

Click to jump to top of pageClick to jump to parent topicSMTPDNSTimeoutRetries

Enables you to configure the number of times the IsDomainNameValid method of the MCFMailUtil class retries to verify that the domain of an email address submitted to the method is valid. If you need to override the system default, manually add this parameter to the [SMTP Settings] section of the PSAPPSRV.CFG file.

See Also

IsDomainNameValid

Click to jump to top of pageClick to jump to parent topicSMTP Further Considerations

Keep in mind the following considerations:

Click to jump to parent topicInterface Driver Options

Set the following parameter for configuring the interface driver for business interlinks.

Note. PeopleSoft Business Interlinks is a deprecated product and these options currently exist only for upgrade compatibility and transition.

Click to jump to top of pageClick to jump to parent topicSCP_LOCALE

Enter the RPS_LOCALE string, which the driver sends to the Supply Chain Planning (SCP) server.

Click to jump to parent topicPSTOOLS Options

You may need to set the following parameters in advanced configurations.

Click to jump to top of pageClick to jump to parent topicEnablePPM Agent

Controls whether the Performance Monitor agent runs and collects performance monitor data. Enter 1 to enable the Performance Monitor agent, and 0 to disable the Performance Monitor agent. This setting overrides the value for this parameter that is set in the database.

See Also

Setting Up the Performance Monitor

Click to jump to top of pageClick to jump to parent topicAdd to CLASSPATH

The Add To CLASSPATH environment variable parameter enables the Java Virtual Machine (JVM) and other Java applications where to find the Java class libraries, including any user-defined class libraries. Because PeopleTools automatically generates CLASSPATH entries for core, delivered class libraries, use this variable to specify additional class libraries that the PeopleSoft software needs to access. To use this parameter, you need to uncomment it in the PSAPPSRV.CFG file.

The PeopleCode API Reference provides details on where you can place custom and third-party Java classes.

See System Setup for Java Classes.

Click to jump to top of pageClick to jump to parent topicJavaVM Options

Specify additional options to be passed to the JVM loaded by the domain's server processes. Separate the options with spaces, for example:

-Xrs -Xmx256m -Xms256m

If the domain will run as a Windows service, you must specify at least the default option, -Xrs.

Note. If you are using the AIX operating system, these options may be required: JavaVM Options=-Xrs -Djava.awt.headless=true -Xcheck:jni.

The JavaVM Options parameter specified in the [PSTOOLS] section specifies global JavaVM options used by every server process in a domain. To override this global value for a particular server process, you can apply custom JavaVM options to individual server processes by adding the JavaVM Options parameter manually to the configuration section for that server process.

JavaVM Options can appear multiple times in a single PSAPPSRV.CFG or PSPRCS.CFG file. While the JavaVM Options value in the [PSTOOLS] section applies to all server processes governed by a particular configuration file, the system only uses the JavaVM Options value in the [PSTOOLS] section for server processes that do not have the JavaVM Options parameter added to its configuration settings section.

For example, if the JavaVM Options parameter has been added to the [PSAPPSRV] section of the PSAPPSRV.CFG file and has been assigned a value, then that value will be used when loading the JVM as a thread of that PSAPPSRV process. If the JavaVM Options parameter does not appear, or has no value, in the [PSAPPSRV] section, then the system uses the value specified in the [PSTOOLS] section when loading the JVM as a thread of the PSAPPSRV process. This applies to any server process: PSAPPSRV, PSQRYSRV, PSAESRV, and so on.

See your JRE documentation for valid JVM options.

Click to jump to top of pageClick to jump to parent topicProxy Host

If the HTTP destination, such as the gateway host, is behind a proxy server for security reasons, enter the distinguished name of the proxy server, as in proxy.oracle.com.

Click to jump to top of pageClick to jump to parent topicProxy Port

Enter the port number on which the proxy server is listening for transmissions. For instance, 80 is a typical default port number.

Click to jump to top of pageClick to jump to parent topicNon Proxy Hosts

Enter a list of the hosts that should be connected to directly, not through a proxy server. Separate the hostnames with a pipe symbol ( | ). You can use an asterisk (*) as a wildcard character to specify a pattern of similar hostnames.

For example, localhost|*.oracle.com indicates that the local host and all hosts with names ending in .oracle.com will be accessed directly.

Note. The length of this string cannot exceed 1024 characters.

Click to jump to top of pageClick to jump to parent topicUsage Monitoring State

Required only if you are enabling the Usage Monitor, which generates system usage metrics using Performance Monitor technology. The usage metrics can then be incorporated with the PeopleSoft Testing Framework to enable you to design more efficient test plans that focus efforts on the elements of the system most used or most affected by and update or upgrade. Usage Monitor must be enabled in the domain before usage metrics can be captured.

Value

Description

0

Usage Monitor is disabled. No usage information will be captured.

1

Usage Monitor is enabled, all users are monitored, and all users are anonymous, as in usage activity is not associated with a particular user ID. Anonymous data is collected unless the user configures the Test Name and Test Case Name fields on the Usage Monitoring page.

2

Usage Monitor is enabled, all users are monitored, and user are not anonymous. That is, usage activity is associated with a particular user ID.

Note. There are other options that need to be enabled within Performance Monitoring in addition to the Usage Monitor State parameter.

See Also

Working With Usage Monitor

Click to jump to top of pageClick to jump to parent topicCharacter Set

Enter the character set to be used as the system locale. This setting is used to convert text to and from UNICODE when using interfaces that do not support UNICODE. Such interfaces include file names, text file contents, and other operating system calls that require non-UNICODE text. This setting controls how files used by the PeopleTools file attachment feature are named. The default values are:

Operating System

Value

UNIX

latin1

z/OS

ccsid1047

Windows

CP_ACP

Note. By default, the parameter is commented out. With no value specified, invalid values or " "(blank value), the system assumes the above default values as the character set.

To override the default, uncomment the Character Set parameter, and select a character set from the following list corresponding to the languages that this application server will process. For example, to set the character set for Simplified Chinese:

Character Set=gb2312

Note. The character set of the application server and the character set of any Microsoft Windows workstations connecting to that application server must match.

The utf8 option is valid only when the locale character set is UTF-8.

See Also

Character Sets in the PeopleSoft Pure Internet Architecture

Click to jump to top of pageClick to jump to parent topicSuppress App Error Box (Microsoft Windows Only)

Enter y to suppress an application error box or message from appearing after an application error occurs. Enter n to view error dialogs and message boxes.

Note. If the system generates an error box for an application server process and this parameter is set to n, Tuxedo can't restart the down process until you close the error box.

Click to jump to top of pageClick to jump to parent topicDbFlags

The following values are valid for the DbFlags parameter:

Value

Description

0

Enable the %UpdateStats meta-SQL construct.

1

Disable the %UpdateStats meta-SQL construct.

2

Ignore the Truncate command for DB2 LUW. Use Delete instead.

4

Disable a secondary database connection (used with the GetNextNumberWithGapsCommit PeopleCode function).

This prevents the creation of a secondary database connection, bundling all SQL into a single unit of work. Without the additional database connection, the database row lock is held for a longer time, reducing concurrency in a multiple-user environment.

Note. Analytic instance processing requires a secondary database connection, so if you're using analytic servers, ensure that this value is not set.

8

Disable a persistent second database connection (used with the GetNextNumberWithGapsCommit PeopleCode function).

This creates a second database connection in each GetNextNumberWithGapsCommit call, then immediately closes the second connection. This keeps the number of database connections to a minimum, but requires each call to create a new database connection on demand.

Note. The performance impact of making a new database connection is significant, especially in high volume user production environments. Don't use this setting without carefully considering its effect.

DbFlags uses a bit mask so that you can specify one or more of these values. You set this parameter to the total of the values that you want to apply. For example, to disable %UpdateStats and ignore the Truncate command, set DbFlags to 3 (setting bits one and two).

The default is value is 1.

See Using PeopleCode in Application Engine Programs, PeopleCode Built-in Functions and Language Constructs.

Click to jump to top of pageClick to jump to parent topicSuppress SQL Error

This option is not available through the PSADMIN interface, but it exists in the PSTOOLS section of the PSAPPSRV.CFG file for small, medium, and large configurations.

For security purposes, this option has a default value of 1 to prevent SQL error details from being displayed to users. Any SQL errors that occur don't display details, but refer users to consult the system log. The details that were in the SQL message are written to the log file. This helps to prevent SQL injection vulnerabilities.

If you want SQL error details to be visible to users, set this property as follows:

Suppress SQL Error=0

Note. In developer configurations, the Suppress SQL Error option doesn't exist in PSAPPSRV.CFG, and the system assumes a value of 0.

See Also

Preventing SQL Injection

Click to jump to parent topicIntegration Broker Options

The following parameters apply to Integration Broker. All of the options for Integration Broker are discussed in detail in the Integration Broker PeopleBooks.

See Also

Administering Messaging Servers for Asynchronous Messaging

Click to jump to top of pageClick to jump to parent topicMin Message Size for Compression

Use this parameter to configure the threshold of a message before the system compresses the message.

Click to jump to top of pageClick to jump to parent topicThread Pool Size

Set the thread pool size used by the SyncRequest PeopleCode event. The Minimum value is 1 and maximum allowable is 20.

Click to jump to parent topicSearch

These options enable you to configure PeopleSoft search. These options are documented in detail in another section of this PeopleBook.

Note. If you do not specify a search configuration type, the system assumes the default configuration based on your operating system.

See Also

Configuring PeopleSoft Search

Click to jump to parent topicSearch Indexes

Use this option to specify the location of all the files pertaining to the search index. Index name is the same as the location. This option is documented in detail in another section of this PeopleBook.

See Specifying the Index Location, Sharing Indexes Between Application Servers and PeopleSoft Process Scheduler.

Click to jump to parent topicPSRENSRV Options

PSRENSRV is a modified web server designed for real time event notification. The primary purpose of PSRENSRV is to publish events to the browser.

See Configuring REN Servers.

Click to jump to top of pageClick to jump to parent topiclog-severity_level

This is the log severity level for the PSRENSRV process. Settings are Error, Warning, Notice or Debug. Default is Warning.

Click to jump to top of pageClick to jump to parent topicio_buffer_size

This is the TCP buffer size when serving content. This should not exceed a value of 65536.

Click to jump to top of pageClick to jump to parent topicdefault_http_port

This is the REN servers http port. The default value is 7180.

Click to jump to top of pageClick to jump to parent topicdefault_https_port

This is the REN servers https port. The default value is 7143.

Click to jump to top of pageClick to jump to parent topicdefault_auth_token

The fully qualified domain name of the server (network domain, not the PeopleSoft application server domain). This value should match the value of the web server's authentication token domain.

Click to jump to parent topicPSPPMSRV Options

PSPPMSRV servers subscribe to performance metrics published by the web service at the PPMI URL (specified in the Performance Monitor administration pages) and insert them into the database. If you select Y when you are asked whether you want Performance Collators configured, then the number of PSPPMSRVs specified in Min Instances=1 will be started. Min and Max instances should be set to the same value, as new PSPPMSRV servers are not spawned on demand.

See Also

Setting Up the Performance Monitor

Click to jump to top of pageClick to jump to parent topicMin Instances

The number of servers started at boot time. This translates to the PSPPMSRV server's –m (min) parameter in the UBB file.

Click to jump to top of pageClick to jump to parent topicMax Instances

The maximum number of servers that can be started. This translates to the PSPPMSRV server's –M (max) parameter in the UBB file.

Click to jump to parent topicSelect Server Process Options

After you enter all of the previous parameter values for the domain, PSADMIN prompts you for the following server process options. You can use these prompts to reduce the number of server processes that start when the domain boots. This, in turn, makes your configuration simpler while conserving system resources.

For instance, if you enter n for any of the following prompts, the corresponding server process (or a set of server processes) is not configured for the domain. If you enter n to all of the prompts, the domain will contain only the minimum required server processes.

Note. You can also enable and disable server processes using the Quick Configure Menu.

Click to jump to top of pageClick to jump to parent topicDo you want the Publish/Subscribe servers configured?

If you want the messaging server processes to be configured and booted, enter y. If you are not implementing the Integration Broker technology for a domain, enter n.

Note. In addition to setting this option, in Integration Broker you must also activate the domain on which the pub/sub server resides before you can use the pub/sub system.

See Administering Messaging Servers for Asynchronous Messaging.

Click to jump to top of pageClick to jump to parent topicMove quick PSAPPSRV services into a second server (PSQCKSRV)?

Enter n if very few clients access the domain and concurrency is not an issue. Enter y to enable the PSQCKSRV in situations where concurrency and optimal transaction throughput are needed.

Click to jump to top of pageClick to jump to parent topicMove long-running queries into a second server (PSQRYSRV)?

If you want all user-generated queries to be initiated by PSQuery and handled by a dedicated server process, enable this option to improve overall performance.

Click to jump to top of pageClick to jump to parent topicDo you want JOLT configured?

The Jolt listener is required to support the PeopleSoft Internet Architecture by enabling transmission between the web server and the application server.

Click to jump to top of pageClick to jump to parent topicDo you want JRAD configured?

JRAD applies to specific configurations only. Accept the default unless you are attempting to configure JRAD for use with the Jolt internet relay.

See Also

Understanding Jolt Internet Relay

Click to jump to top of pageClick to jump to parent topicDo you want WSL Configured?

Configures the Workstation Listener for Development Environment (Windows) workstation connections.

Click to jump to top of pageClick to jump to parent topicDo you want to enable PeopleCode Debugging (PSDBGSRV)?

Enter y to debug PeopleCode programs with the current domain.

Click to jump to top of pageClick to jump to parent topicDo you want Event Notification configured (PSRENSRV)?

Select Y to start the PSRENSRV servers.

See PSRENSRV Options.

Click to jump to top of pageClick to jump to parent topicDo you want MCF Servers configured?

Select Y to start the Multi Channel Framework servers.

See Configuring PeopleSoft MCF Queues and Tasks.

Click to jump to top of pageClick to jump to parent topicDo you want Performance Collators configured?

If the domain is servicing a Performance Monitor database, select Y to start the PSPPMSRV servers.

See PSPPMSRV Options.

Click to jump to top of pageClick to jump to parent topicDo you want Analytic Servers configured?

Configures analytic servers to run in the domain to process Analytic Calculation Engine requests and to perform optimization processing.

Click to jump to top of pageClick to jump to parent topicDo you want Domains Gateway configured?

Enable this option of you are configuring a remote , or external, search server to which this domain will send search requests. That is, if you are configuring a Type-3 search option for an application server domain, you need to enable the domains gateway on the application server domain to a communication connection between the application server and its remote search domain.