78727688 Data Stage Admin Guide

DataStage Admin Activities

March 2010

Wipro Technologies@ Office Depot

© Wipro Ltd - ConfidentialApril 7, 2023 1

Confidential © Copyright 2007 Wipro Ltd

Content

April 7, 2023 © Wipro Ltd - Confidential 2

• Purpose & Scope• Start/Stop DS Engine• To Check the status of DS Engine• To Release Job locks• To Delete Jobs• To Delete Datasets• To Restore a Dataset• System Monitoring Tools

– iostat– vmstat

• To View Who Last Modified a Job• DataStage Housekeeping • Add Environment Variables• Suggested Environment Variables For Reporting • Import DataStage Jobs From Command Prompt• Export DataStage Jobs From Command Prompt• Compiling Jobs• Message Handler• To Configure/Test DataBase connectivity

– Oracle– Other Databases


Purpose & Scope


• Purpose

Purpose of this Presentation is to give some knowledge on DataStage Admin Activities performed on Server side and DataStage Client Side.

• Scope

This document will educate you about DataStage repository database (UV). This presentation doesn’t cover all the properties of UV Database. This document will educate you about Configuring and trouble shooting

the database connectivity for DataStage.


Start/Stop DS Engine


From time to time you may need to stop or restart the DataStage engine, for example, when you want to shut down the server.

A script “uv” is provided for these purposes.

su to dsadm (su – dsadm) and thenIssue appropriate stop or start command below.

To stop the server engine, use:

# dshome/bin/uv -admin –stop

To restart the server engine, use:

# dshome/bin/uv -admin –start


To Check the status of DS Engine


The following command is used to check the status of the DS Engine. $ uv -admin –info

The following data will be displayed

$ uv -admin -infoDetails for DataStage Engine release 7.5.2.1 instance "ade"================================================================Install history : Installed by root (admin:dsadm) on: Feb 03 2005 14:18:21 : Upgraded by root (admin:dsadm) on: Sep 21 2006 17:20:25 : Upgraded by root (admin:dsadm) on: Sep 21 2006 19:53:45 : Upgraded by root (admin:dsadm) on: Sep 21 2006 20:57:20 : Upgraded by root (admin:dsadm) on: Sep 22 2006 10:17:17 : Upgraded by root (admin:dsadm) on: Sep 22 2006 11:53:46 : Upgraded by root (admin:dsadm) on: Sep 22 2006 14:32:55Instance tag : adeEngine status : Running w/active nlsEngine location : /ascential/Ascential/DataStage/DSEngineBinary location : /ascential/Ascential/DataStage/DSEngine/binImpersonation : EnabledAdministrator : dsadmAutostart mode : disabledAutostart link : noneStartup script : /ascential/Ascential/DataStage/DSEngine/sample/ds.rcCache Segments : 0 activeUser Segments


To Release Job locks


There are Three methods to unlock the DataStage jobs

– Using DataStage Administrator Tool.– Using UV Utility – Using DataStage Director


Unlock jobs -Using DataStage Administrator Tool


• Log into Administrator (as dsadm)• Open the Command line for the project• Execute “LIST.READU EVERY”• Identify the values for INODE and USER columns for the job for which the locks

need to be released.• Execute UNLOCK INODE <inodenumber> ALL• UNLOCK USER <user number> ALL


Unlock jobs -Using UV Utility


• Logon to Unix• Su to dsadm• Go to a corresponding project• Type “UV”• Type “DS.TOOLS”• Select the option “4” (4. Administer processes/locks >>)• Again Select the option “4” (4. List all locks)• Identify the values for PID columns for the job for which the locks need to be

released.• Then select the Option “7” (7. Clear locks held by a process)• Give the PID value here(Enter pid#=) and press Enter.


Unlock jobs -Using DataStage Director


– Open DataStage Director– Go to the Job – Cleanup Resources– In the Job Resources window,– Select “Show All” (Processes)– Find your User Name and click on “Logout”


To Delete Jobs


The DataStage jobs can be deleted either via DS Administrator or the TCL prompt.

1. In Administrator select the project and run this command:DELETE DS_JOBS <job name>

2. From the TCL prompt you will need to do this:$DSHOME/. ./dsenv./bin/uvshLOGTO <project name>DELETE DS_JOBS <job name>

Using TCL, you could automate this via script. Also developers can delete the jobs through Designer Client.

Note: The deleted jobs can be restored only with the disk backup


To Delete Datasets


The Datasets in DataStage can be deleted using following two methods,

1. Using orchadmin2. Using DataSet Management

Using orchadmin

The orchadmin command is used to delete the datasets$APT_ORCHHOME/bin/orchadmin delete <dataset name>For Example orchadmin delete

/detld1/etl/Ascential//DataStage/Projects/Staging/EDPRO_20100225_129.ds

Using DataSet Management

The DataSet Management utility is available in DataStage Manager and Director. Using this we can delete the datasets.


To Restore a Dataset


The dataset can be restored from a disk backup

• Restore the header of the dataset, which will be found in the directory where

specified in the job.

• Restore the detail records of the dataset, Which will be found in the resource

directory specified in the config file used to run the job.


System Monitoring Tools


iostat

iostat is useful for examining the throughput of various disk resources. If one or more disks have high throughput, understanding where that throughput is coming from is vital. iostat can also help a user to determine if there is excessive IO for a specific job.

The specifics of iostat output vary slightly from system to system. Here is an example from a Linux machine which shows a relatively light load:

(The first set of output is cumulative data since the machine was booted)

$ iostat 10Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtndev8-0 13.50 144.09 122.33 346233038 293951288

… every N seconds (10 in the command line example) iostat outputs:Device: tps Blk_read/s Blk_wrtn/s Blk_read Blk_wrtndev8-0 4.00 0.00 96.00 0 96


System Monitoring Tools


vmstat vmstat is useful for examining system paging. Ideally, a EE flow, once it begins

running, should never be paging to disk (si and so should be zero). Paging suggests EE is consuming too much total memory.

$vmstat 1 procs memory swap io system cpu r b w swpd free buff cache si so bi bo in cs us sy id 0 0 0 10692 24648 51872 228836 0 0 0 1 2 2 1 1 0

vmstat produces the following every N seconds: 0 0 0 10692 24648 51872 228836 0 0 0 0 328 41 1 0 99

mpstat will produce a similar report based on each processor of an SMP


To View Who Last Modified a Job


• Logon to Unix• Su to dsadm• Go to a project folder• Type “UV”• Issue the following SQL

SELECT INSTANCE, DTM, MODIFIER FROM DS_AUDIT WHERE CLASS = '2' AND INSTANCE LIKE '%MyJob';


DataStage Housekeeping


There is very little housekeeping that can be done from the OS level.

– Clearing /tmp folder – Clearing the &PH& directories– Delete the temporary files, Data Sets or File Sets. It is advised to

do this in the application side itself.


DataStage Housekeeping


Before clearing the &PH& directories make sure ever one logged out of the datastage.

– To check the user connections use the command “uv –admin –info” .– From any project directory– . $DSHOME/dsenv– $DSHOME/bin/uv– LOGTO UV– RUN ADMIN_BP REMOVE.SHM– Or $DSHOME/bin/dsdlockd -p

Run the following command for each project.

– LOGTO (Project Name)– CLEAR.FILE &PH&


Add Environment Variables


The following are the steps to add the environment variables in DataStage

– Logon to DataStage Administrator – Choose the Project– Go to Environment– Select User Defined tab to add the new used defined variables.– Select the other options to change the values for the existing

variables.


Suggested Environment Variables for Reporting


The following environment variables are concerned with various aspects of WebSphere® DataStage® jobs reporting their progress.

– APT_DUMP_SCORE– APT_ERROR_CONFIGURATION– APT_MSG_FILELINE– APT_PM_PLAYER_MEMORY– APT_PM_PLAYER_TIMING– APT_RECORD_COUNTS– OSH_DUMP– OSH_ECHO– OSH_EXPLAIN– OSH_PRINT_SCHEMAS


Import DataStage Jobs from Command Prompt


Using DS_IMPORTDSX Command

This command runs within the DS engine using dssh. It canimport any job executables found within specified DSX files. Run it asfollowing,

1. CD to the project directory on the server:2. Run the dssh shell: ..\..\engine\bin\dssh3. At the dssh prompt enter the DS_IMPORTDSX command

The DSXimport command is as follows,

DS_IMPORTDSX filename [[-OVERWRITE] -JOB[S] * | jobname …] | [-LIST]


Import DataStage Jobs from Command Prompt


The following are the arguments,

<ilename> The name of the DSX file containing the components to import.

<OVERWRITE> Specify this to overwrite any existing executables of the same name.

<JOB[S]> Specify one or more job executables to import. <LIST> Specify this to list the executables in a .DSX file rather than import them.

If you don’t specify a JOB or a LIST argument, the default is to import all job executables in the specified file.


Export DataStage Jobs from Command Prompt


Using dscmdexport command

The dscmdexport command used to export the datastage jobs from Command prompt,

dscmdexport /H hostname /U username /P password /O omitflag project pathname /V

The arguments are as follows: <hostname> Specifies the DataStage Server from which the file will be

exported. <username> The user name to use for connecting to the DataStage Server. <password> The user’s password.


Export DataStage Jobs from Command Prompt


<omitflag> Set this to 1 to omit the username and password (onlypossible if you are connected to the DataStage Server via LANManager).<project> Specify the project to export the components from.<pathname> The file to which to export.<V> Use this flag to switch the verbose option on.For example, the following command exports the project dstage2from the R101 to the file dstage2.dsxdscmdexport /H R101 /O 1 dstage2 C:/scratch/dstage2.dsxMessages from the export are sent to the console by default, but canbe redirected to a file using '>', For example:dscmdexport /H R101 /O 1 dstage99 c:/scratch/project99.dsx /V >

c:/scratch/exportlog

You can simply type dscmdexport at the command prompt to get help on the command options


Compiling Jobs


There is a program on the Client side called dscc.exe. This resides in the Clients\Classic folder where the Designer and Director reside. This is used to Compile the jobs.

C:\IS81CL\Clients\Classic>dscc /?/? Show program usage/bo (value) BuildOp to compile, * = All BuildOps, \folder\* = folder/d (value) ASB Domain Name/f Use the Force Compile/h (value) Name of Host to attach to/j (value) Job to compile, * = All Jobs, \folder\* = folder/jt (value) Job Type To Compile -1 (Default) = All Types 0 = Server 1 = Mainframe 2 = Sequence 3 = Parallel/mfcgb (value) Mainframe Code Gen base directory location/mful (value) Mainframe Job Upload Profile/ouc Only Compile Uncompiled Objects/p (value) Passwordproject Project To Attach to/qs (value) QS objects to provision * = All QS objects, \folder\* = folder/qspa Provision all dependant QS objects/r (value) Routine to compile, * = All Routines, \folder\* = folder/rd (value) Report Directory and Name/rt (value) Type of report to produce : X = xml, T = txt - default/u (value) User Name

Also the jobs can be compiled from DataStage Designer Client or from Manager through “Run Multiple Job Compile” Option.


Message Handler


The Message Handler is used to suppress the Warnings.

• Log into the Director• Select the job• And then view the logs• Select the warning• Right click on the Warning message• A popup will display• Select “ Add Rule to the Message Handler”• Now choose the “Add rule to chosen message handler”• And click on the Button “ Add Rule”


To Configure/Test DataBase Connectivity


For Oracle

Configure Tnsnames.ora file, Which defines databases addresses for establishing connections. This file normally resides in the ORACLE HOME\NETWORK\ADMIN directory

Add the following entry in your TNSNAMES.ORA file and change the values shown in brackets to suit your environment:

<addressname> = (DESCRIPTION = (ADDRESS_LIST = (ADDRESS = (PROTOCOL = TCP)(Host = <hostname>)(Port = <port>)) ) (CONNECT_DATA = (SERVICE_NAME = <sid>) ) )

This can also be done through LDAP

http://www.orafaq.com/wiki/ORACLE_HOME




Tnsping

TNSPING is a utility in the ORACLE HOME/bin directory used to test if a SQL*Net connect string can connect to a remote listener (check if the socket is reachable).

Note: This utility only tests if the listener is available. It cannot tell if the databases behind the listener is up or not

For Example

$ tnsping myDB

TNS Ping Utility for Linux: Version 10.2.0.1.0 - Production on 24-MAY-2007 08:55:13Copyright (c) 1997, 2005, Oracle. All rights reserved.Used parameter files:/app/oracle/product/10.2.0/db_1/network/admin/sqlnet.ora

Used TNSNAMES adapter to resolve the aliasAttempting to contact (DESCRIPTION = (ADDRESS = (PROTOCOL = TCP)(HOST = localhost)(PORT = 1521)) (CONNECT_DATA = (SERVER =

DEDICATED) (SERVICE_NAME = MyDB)))OK (10 msec)


TO Configure/Test DataBase Connectivity


When using an LDAP (OID) server - NAMES.DIRECTORY_PATH=(LDAP) in sqlnet.ora:

$ tnsping myDB

TNS Ping Utility for Solaris: Version 9.2.0.7.0 - Production on 10-DEC-2007 15:05:50Copyright (c) 1997 Oracle Corporation. All rights reserved.Used parameter files:/app/oracle/product/10.2.0/db_1/network/admin/sqlnet.ora

Used LDAP adapter to resolve the aliasAttempting to contact (DESCRIPTION = (ADDRESS = (PROTOCOL = TCP)(HOST =

localhost)(PORT = 1521)) (CONNECT_DATA = (SERVER = DEDICATED)(SERVICE_NAME = MyDB)))

OK (300 msec)Retrieved from "http://www.orafaq.com/wiki/Tnsping"Categories: Network services | Utilities




The SQLPLUS is used to check the databases behind the listener is up or not

The following is the syntax,

sqlplus username/password@servicename

http://www.adp-gmbh.ch/ora/network/net_service_name.html




Other Databases

• Add an entry in odbc.ini file.• Every DSN to which the application connects must have an entry in this file.

The following example shows the configuration for a DSN called Informix 9

[Informix9] Driver=/work/informix/lib/cli/iclis09b.so Description=Informix 9.x ODBC Driver LogonID=user1 pwd=tigress4 Database=odbc_demo ServerName=my_server


TO Configure/Test DataBase Connectivity


The datadirect example program is used to check where the database is in relation to the DS Server.

The following are the steps used to check this,

$cd /export/home/752NLS/Ascential/DataStage/DSEngine$. ./dsenvThen cd to the branded_odbc/example directory $cd /export/home/752NLS/Ascential/DataStage/branded_odbc/exampleand run the example program $./example

When prompted put in the odbc entry name and the credential similar to this below sample:

$ ./example./example DataDirect Technologies, Inc. ODBC Example Application.Enter the data source name : <yourdatasourcename from .odbc.ini>Enter the user name : <yourusername>Enter the password : <youruserpassword>

And then you should see the below (or otherwise some type of error)

Enter SQL statements (Press ENTER to QUIT)SQL>Exiting from the Example ODBC program

Thank You


78727688 Data Stage Admin Guide

Documents

datastage

release job

export datastage

import datastage

dscmdexport

environment

message handler

ds engine