Exporting assets from the source system (Watson Knowledge Catalog)

On your source system, export the data that you want to migrate to Cloud Pak for Data.

Required roles in the source system
To export assets from Information Server, you must have the Suite Administrator and Common Metadata User roles.
Order when exporting asset types
Export the asset types that you want to migrate to the target system in any order. However, it is important to import assets in a specific order. Therefore, the export procedures are also presented in the same order in which the import must be completed.
Location of the istool command
In many cases, you use the istool command to export assets. The command is in the installation_directory/Clients/istools/cli directory, where installation_directory is for example /opt/IBM/InformationServer.
Dividing export files into chunks
When you use the istool command to export assets, you can specify the -maxArchiveSize option to improve the performance. This option creates additional archive files when the size of the export file exceeds the specified number of megabytes. In the example commands, this parameter is set to 500, so the exported files don’t exceed the limit of 500 MB. This is a recommended value for the migration.

See the following sections for the export instructions.

Bulk export of asset groups by using the export -all command

You can use the istool export -all command to export many asset types at the same time. Alternatively, if you’d rather export asset groups one by one, you can use commands specific to the asset group. Exporting asset groups one by one might be the preferred solution when you want to use additional options with the export commands.

The following table provides a list of asset groups which are exported by the istool export -all command, the alternative commands for each asset group, and links to resources with more information about each command.

Table 1. Istool export -all details
Asset group Alternative istool command More information (Information Server documentation)
Common metadata assets:
  • Implemented data resources
  • Business intelligence (BI) assets
  • Physical data model assets
  • Data connections
  • Contract libraries
 export -commonmetadata
Options automatically used by the export -all command:
  • -includeContactAssignment
  • -includeAnnotations
 Export command for common metadata assets
Data quality assets:
  • Workspaces
  • Data sets
  • Data classes
  • Data rules
  • Data rule definitions
  • Rule sets
  • Rule set definitions
  • Quality rules
  • Analysis results: column analysis, data quality analysis, data rule history, rule set history, overlap analysis, relationship analysis, cross-domain analysis, multiple column primary key analysis
  • Metrics
 export -ia
Options automatically used by the export -all command:
  • -tablelevel
  • -includeReports
  • -includeDataClasses
  • -includeResultHistory
  • -includeCommonMetadata
  • -includeProjectRoleAssignments
 Export command for InfoSphere Information Analyzer assets
Extension mapping documents and extended data sources export -igc Extension mappings export command
InfoSphere Metadata Asset Manager import areas and data connections (staging area assets are not exported) export -imam Export command for import areas
InfoSphere Streams assets export -streamsEndpoint Export command for InfoSphere Streams assets
Export all assets at the same time
To export all assets at the same time, run this command:
./istool.sh export -dom host:port -username username -password password
-archive "/tmp/iisassets.isx" -maxArchiveSize 500 -all -Xmx4096M
The -Xmx option specifies the max heap size to use when the command runs. For more information about the command, see the istool export -all topic in the Information Server documentation.
Export each asset group individually
To export each asset group individually, run these commands.
Common metadata assets
./istool.sh export -dom host:port -username username -password password -archive "/tmp/commonmetadata.isx" -commonmetadata 'asset_identity_string -includeAnnotation -includeContactAssignment' -maxArchiveSize 500 -Xmx4096M
For information about the value of asset_ identity_ string, see the Common metadata asset types and identity strings for the command line topic in the Information Server documentation. For example, to export all assets of the specified types, run this command:
./istool.sh export -dom host:port -username username -password password -archive "/tmp/commonmetadata.isx" -commonmetadata '/*/*.db /*/*.dcn /*/*.did /*/*.fd /*/*/*.fdr /*/*.pm /*.srv /*.cl /*/*/*.fl -includeAnnotation -includeContactAssignment' -maxArchiveSize 500 -Xmx4096M
Data quality assets
./istool.sh export -dom host:port -username username -password password -ar /tmp/dataquality.isx -ia ' -projects="*" -includeDataClasses -includeResultHistory -includeProjectRoles -includeReports -tablelevel' -maxArchiveSize 500 -Xmx4096M
Extension mapping documents and extended data sources
./istool.sh export -dom host:port -username username -password password -archive "/tmp/igcext.isx" -igc '/*.eds, /*/*.emd' -maxArchiveSize 500 -Xmx4096M
InfoSphere Metadata Asset Manager import areas
./istool.sh export -dom host:port -username username -password password -archive "/tmp/importarea.isx" -imam '-importArea "*"' -maxArchiveSize 500 -Xmx4096M
If you want to export staging area assets as well, use the -includeStagingAreaAssets option, as in:
./istool.sh export -dom host:port -username username -password password -archive "/tmp/importareastaging.isx" -imam '-importArea "*"' -includeStagingAreaAssets -maxArchiveSize 500 -Xmx4096M
InfoSphere Streams assets
./istool.sh export -dom host:port -username username -password password -archive "/tmp/streams.isx" -streamsEndpoint '-ep *' -maxArchiveSize 500 -Xmx4096M

Data classes

Cloud Pak for Data supports only one regular expression for a data class. If you have data classes with an additional regular expression defined, you must split these data classes in your source system so that each has only one regular expression. If you don’t split such data classes, only the main regular expression is migrated. To export data classes, run this command:
./istool.sh export -dom host:port -username username -password password -archive "/tmp/dataclasses.isx" -dca '/*.dc /*.als -inccls -incca -incannot' -maxArchiveSize 500 -Xmx4096M
The -Xmx option specifies the max heap size to use when the command runs. For more information about the command, see the Export command for data class and data collection analysis assets topic in the Information Server documentation.

Custom attribute definitions

To export custom attribute definitions, use the Information Governance Catalog user interface. Complete these steps:
  1. Log in to Information Governance Catalog classic. Access it by using this URL:
    https://source-host-name/ibm/iis/igc/
  2. Go to Administer > Custom attribute definition.
  3. Select all custom attribute definitions and, from the menu, select Export.
  4. Save the XML file.

Bundle assets (open IGC)

You can’t export bundles from Information Server, but you can register the same bundles in Cloud Pak for Data. For more information, see Bundle assets (open IGC) in the import section. You can export bundle assets and flows. Complete these steps:
  1. Open the Information Governance Catalog REST API by using this URL:
    https://source-host-name/ibm/iis/igc-rest-explorer/
  2. Use the following commands to export bundle assets and flows. You must repeat the procedure for all bundles and flows.
    • To export bundle assets, expand the bundles section and use the GET/bundles/assets command. Find the name of a bundle by using GET/bundles command and provide it in the family parameter. Click Try it out. Save the XML file with the results.
    • To export flows, expand the flows section and use the GET/flows/ command. In the asset parameter, specify the flowUnit IDs. If you use more than one value, separate them with commas. Click Try it out. Save the XML file with the results.
For more information about REST API commands, see the Adding asset types, their assets, and their data flows into the catalog technote.

InfoSphere DataStage and QualityStage assets

To export InfoSphere DataStage and QualityStage assets, run this command:
./istool.sh export -dom host:port -username username -password password -archive "/tmp/project_assets.isx" -ds '"dsServer/project/folder/asset_name.suffix"' -maxArchiveSize 500 -Xmx4096M
The -Xmx option specifies the max heap size to use when the command runs. The "dsServer/project/folder/asset_name.suffix" option specifies the paths of assets to be added to the exported file. For example, to export all assets in the project dstage1, run this command:
./istool.sh export -dom host:port -username username -password password -archive "/tmp/dstage1_assets.isx" -ds 'IIDEMO5/dstage1/*/*.*' -maxArchiveSize 500 -Xmx4096M
Run this command for each project that you want to export.

For more information, see the Export command for InfoSphere DataStage and QualityStage assets and Asset paths for InfoSphere DataStage and QualityStage assets topics in the Information Server documentation.

Glossary assets

Glossary assets include terms, categories, information governance rules, information governance policies, and labels. You must export them to an XML file.

Export glossary assets by running this command:
./istool.sh glossary export -dom host:port -username username -password password -filename "/tmp/glossaryassets.xml" -format XML -allcategories -allpoliciesrules -includeassignedassets -includestewardship -includelabeledassets -Xmx4096M
The -Xmx option specifies the max heap size to use when the command runs. For more information about this command, see the Glossary export command topic in the Information Server documentation.

Lineage filters and report templates

To export lineage filters and report templates, complete these steps:
  1. In Information Governance Catalog classic, go to Administration > Lineage Management.
  2. To export lineage filters:
    1. Open Manage Lineage Filters.
    2. Select the filters that you want to export.
    3. From the menu, select Export.
    4. Save the file.
  3. To export lineage report templates:
    1. Open Manage Lineage Report Templates.
    2. Select the templates that you want to export.
    3. From the menu, select Export.
    4. Save the file.

Automation rules

Export automation rules by running this command:
/opt/IBM/InformationServer/ASBServer/bin/IAAdmin.sh -user username -password password -url https://host:port -getDataQualityConfigurationRule
For more information about the command, see the Commands to import, export, and delete automation rules topic in the Information Server documentation.

Analysis database, automated discovery results, and data rule and rule set run history

Run this procedure for each data type separately. Replace db_name and schema_name with the following values:
Data type db_name schema_name
Analysis database IADB IAUSER
Automated discovery results XMETA AUDIT_TRAIL
Data rule and rule set run history XMETA ASSET_RELATION
This procedure is specific to Db2 databases.
Complete these steps:
  1. In the source system, log in to the database as the db2inst1 user and set up the environment to run Db2 commands by running this command:
    . ~/sqllib/db2profile
  2. Create the database directory in the /tmp directory.
    mkdir /tmp/db_name
chmod 777 /tmp/db_name
cd /tmp/db_name
  3. Export database assets by running this command:
    db2move db_name export -aw -l lobs -sn schema_name

Quick scan results

To export quick scan results, complete these steps:
  1. Log in to the solr pod on the source system:
    kubectl exec -it solr-0 bash
  2. Create temporary directories:
    mkdir /var/solr/data/solr_analysis_backup
    mkdir /var/solr/data/solr_discovery_backup
  3. Run the following backup commands:
    curl -k 'http://localhost:8983/solr/analysis/replication?command=backup&location=/var/solr/data/solr_analysis_backup'
    curl -k 'http://localhost:8983/solr/discovery/replication?command=backup&location=/var/solr/data/solr_discovery_backup'
  4. To display the status of the backup process, run these commands:
    curl -k 'http://localhost:8983/solr/analysis/replication?command=details&wt=xml'
    curl -k 'http://localhost:8983/solr/discovery/replication?command=details&wt=xml'
  5. Copy the solr_analysis_backup and solr_discovery_backup directories from the solr pod to the tmp directory on the source system.

Data connections with mappings

You must export data connections separately so that you can later import them with a mapping document. The mapping document is used to specify the source and target host names. It is required when the source and target host names differ.

To export data connections, run this command:
./istool.sh export -dom host:port -username username -password password -archive "/tmp/dataconnections.isx" -cm '/*/*.dcn' -maxArchiveSize 500 -Xmx4096M
The -Xmx option specifies the max heap size to use when the command runs. For more information about this command, see the Export command for common metadata assets topic in the Information Server documentation.