Using the IBM Business Automation Workflow data collector

Edit online

Use the IBM Business Automation Workflow data collector, which is installed with IBM Business Automation Workflow, to gather some of the standard information that is needed to debug IBM Business Automation Workflow issues. This command-line tool can gather log file or configuration information.

About this task

The IBM Business Automation Workflow data collector gathers various product information, compresses it into a .zip file, and optionally sends the compressed file to IBM for a case. You can also use the tool to compress the data without sending it to IBM. The command is 
profile_root/bin/bpmdc
Where profile_root is the base profile directory for the profile from which you want to get data. For example:
 C:/MyProfileDirectory/bin/bpmdc -prefix=TS123456789
You might use the IBM Business Automation Workflow data collector tool in these situations:
  • After an issue occurs, you can use the basic bpmdc command to compress the log files for the profile into a .zip file.
  • You can re-create an issue by setting a trace string using the administrative console or the wsadmin command, and then run the bpmdc command to collect the logs and send them to IBM.
  • After opening a case, you can use the bpmdc -prefix=caseNumber command with the most relevant type options to prepend the caseNumber number to the collected data file.
  • IBM Support provides a specific command or set of commands to run for your environment that obtains the needed data.
  • IBM Support provides a trace string that must be gathered with the commands. With trace enabled, you can re-create the issue or wait for a recurrence before running the provided commands to gather and upload the data to IBM.
The following table lists the command line parameters.
Table 1. Command line parameters
Command line parameter Default value Description
prefix None prefix | None | The case number that is associated with the data, such as TS123456789.
outputDirectory install_root/BPM/isadc The path to the directory where the output collection .zip file will be stored.
type General The type of collection to run. The valid values are General, Install, Dump, or TableDump.
config no With a general collection, when you set this parameter to yes, the config directory is included. You can set this value to yes or no.
help No values When you include this parameter with the command, a description of the command options is displayed.
The following table lists the server connection options required for -type=Dump.
Table 2. Server connection options
Command line parameter Default value Description
server None The server from which you want to collect JVM dumps.
soapport None The SOAP port for the server.
user None The user with wsadmin access to the server to which you are connecting.
password None The password for the user.
The following table lists the dump options available for -type=Dump.
Table 3. Dump options
Command line parameter Default value Description
dumptype Thread The type of Java™ virtual machine (JVM) dump to trigger:
  • Thread - Javacore or thread dump.
  • Heap - Heap dump.
  • System - System dump, with JExtract run on it before the upload. This type can take some time, depending on the heap size.
count 1 The number of JVM dumps to trigger during the collection.
interval 0 The number of JVM dumps to trigger during the collection.
The following table lists the table dump options available for -type=TableDump.
Table 4. TableDump options
Command line parameter Default value Description
dbname The name of the database or instance to which you want to connect for the TableDump type. It is optional for Oracle databases.
dbUser

dbPassword

 The user name and password for accessing the database tables for the TableDump type. The user name and password are optional because the data collector by default pulls the details from the data source entities that are defined for the workflow server. The user name and password might be required if the system has a unique permission set. If the database being used is an Oracle database, the user name and password are required.
component The option to specify the workflow component name for the TableDump type.
queryFile The option to specify the custom query file path that is used to specify tables for the TableDump. It cannot be used together with the component or tableNames option. Ensure that you have the correct database schema name for the tables that you are collecting. A sample query file is found at the following location:

BPM/isadc/util/components/Query.txt
tableNames The option to specify tables for the TableDump type. It cannot be used together with the queryFile or component option. You can include multiple table names that are separated by a comma and enclosed in back slash and double quotation mark characters. For example:

-tableNames=\"Table1,Table2\"
delim , The option to specify a delimiter for an exported comma-separated value (CSV) for the TableDump type. The character must be enclosed in double quotation marks (" ").
clusterName The option to specify the cluster name of the environment for the TableDump type.

You can run the command without any parameters, in which case the general collection is run and the log directory is compressed into a .zip file.

The output file name has the following format: prefix.ProfileName_type_timeStampValue.zip

The data gathered is based on the profile where the command is run. If it is run from the installation directory, the default profile is used.
Note: Be sure to run the command from the profile from which you want to gather data.
General collection
The General collection type gathers the following information:
  • Profile logs directory (profile_root/logs). The collected files include all of the server log files and traces under this profile directory.
  • The results of the versionInfo -maintenancePackages command.
  • Optionally, the profile_root/config directory. This directory is collected only when the -config=yes option is used.
Note: The limitation with this collection is that it does not gather log files and traces if it goes to a custom directory outside the profile logs directory. It is not recommended that you change the log file locations from the default, because it makes it more difficult for others to find them.
Install collection
The Install collection type gathers the following information, which is listed in the technote Collect troubleshooting data for installation or upgrade problems with IBM Business Process Manager (BPM) products:
  • Any temporary launchpad log files
  • Installation Manager data directory
  • The results of the versionInfo -maintenancePackages command
  • Installation logs directory
  • Profile logs directory
  • Database upgrade script log files
  • Various configuration files that are related to the installation process
JVM Dump collection
The Dump collection type triggers various types of JVM dumps on a target server within a profile before gathering the general collection and generated dumps.

With the provided server connection options, the command will connect to the server using the SOAP port to make a wsadmin connection with the user and password. Depending on the dump type, it will trigger various JVM dumps until the count number is reached. The dumps are triggered a set number of seconds apart based on the interval that is set.

The Dump collection gathers general collection files such as /logs and version information, in addition to dump files such as javacores, heap dumps, and system dump .zip files.

Note: The server connected to must be the one you want to gather data from.
TableDump collection
The TableDump collection type triggers database dumps on a target server within a profile before gathering the general collection. Using the provided database name and other optional properties, the command will connect to the server using the database username and password, as specified in the workflow server data sources in the server configuration files. The available optional properties are for selecting specific tables or groups of tables, user name and password, and the delimiter character for the generated table dump.