Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

Installation

...

  • Configure proper license key

    • Based on the client’s source, generate the right key using the License Key generator: Oracle EPM Planning / Oracle EPM Financial Consolidation and Close.
      Path to generator: \\cxo-fileserv\Data\Internal Tools\(not-released) LicenseKeyGenerator for B Licenses (with Oracle Cloud EPM adapter)

  • Create a new source system in the SSM for the correct source type: Oracle EPM Planning / Oracle EPM Financial Consolidation and Close. This should create:

    • Fact database

    • SSAS cube

  • Run configuration database creation command: e.g.
    .\CXO.Adapter.OracleCloudEpm.Cli.exe create-configuration-db server-name:cxo-dev-sql02 database-name:oracleEpmConfigTest authentication-type:Proprietary username:cxo password:somePassword
    Parameters:

    • server-name: name of the server on which the database should be created

    • database-name: name of the database to be created

    • authentication-type: windows / proprietary

    • username: SQL user name (if authentication-type == proprietary)

    • password: SQL user password (if authentication-type == proprietary)

...

  • Make sure we have an account that we can use to connect from the adapter to the Oracle system (We probably need user with admin rights for it)

  • Configure metadata export job

  • Configure data export job

...

Configure metadata export job

For Gjensidige, there should already be a metadata job defined for the old adapter. We could reuse that, if there are no weird values.

Steps to perform:

  1. Go to Overview screen

...

3. On the export metadata screen you will see the existing CXO job, might be others as well.

a. Review existing job, if there are no weird values we can reuse it. (all delimiters should be same)b. If needed, create new job. See two screenshots below

...

...

b. select all dimensions which should be extracted

...

Configure data export job

...

job

...

  1. Go to the Overview screen

  2. Choose Export Data in the dropdown on the right side of the screen

  3. Create a new data job

    Image RemovedImage Added
  4. Specify the following values:

    1. Location: Outbox

    2. Pick right cube

    3. Choose a delimiter, preferably set it to “Other” and use “|”

    4. Smart Lists are not used currently

    5. Dynamic Members: Exclude

    6. Decimals: none (or discuss with Gjensidigethe customer)

    7. Slice definition: take Account on rows, Period on columns, set default slice values for all dimensions incl. POV. Note: this configuration will not be used in the actual data extraction, as it will be input to the job API call.

...

Order of POV dimensions is important – we parse data extraction file using it. To figure it out fill in at first in a random order, finish configuration and return to the next step after you have successfully run an extraction. (you will see in logs that data extraction is finished successfully but there would be worning warning stating that 0 records were mapped)

...

Selector functions are not supported yet (e.g. Children(YearTotal)) there is no worning warning for this so make sure there are not included in Data Slices. Use multiple members to replace such functions.

Example:

...

    ...

    • For Gjensidige: Check the old PBCS configuration for connection info, dimension mappings etc.

    • Set up connection information (see old PBCS)

    • Map the Oracle dimensions to CXO dimensions (see old PBCS)

    • Set default properties for metadata extraction: delimiters, job name, aliases

    • Configure data extraction settings

      • Set default properties: delimiters, job name

      • Set file structure, matching the configuration in the Oracle EPM system

        • Put it in the database in Source table: row, then column, then POV in order

      • Set data filters

        • 1 member expression per mapped dimension in DataFilters table, excluding period, year and scenario

    ...

    • Run metadata extraction

      • Command:
        CXO.Adapter.OracleCloudEpm.Cli.exe extract-metadata source:<source-id-here>

      • Review hierarchies, add metadata filters for members that should be excluded from extraction

    • Run data extraction

      • Command:
        CXO.Adapter.OracleCloudEpm.Cli.exe extract-data source:<source-id-here> period:Jan year:FY2021 scenario:Forecast

    • Use Run now button to extract data and metadata

    ...

    • You can see the progress of the extraction in the logs tab

    • After successful extraction Process cube from SQL Server Management Studio (this is temporary solution. This process will be automated in the future)

    Logging

    • The log level can be set in appsettings.json in "LogLevel" property (Both Web project and CLI project have own appsettings.json file)

    • Location of the log files:

      • CLI command {CXO-Installation-Dir}\Oracle Cloud EPM Adapter\Extraction\logs

      • Web application {CXO-Installation-Dir}\Oracle Cloud EPM Adapter\Web API\logs

    • Log files are split into two file types

      • nlog-all -contains cxo defined logs and logs generated by other Microsoft libraries. Usually it is more verbose

      • nlog-own-detailed -contains logs defined by cxo with more detailed information then info in “nlog-all“ and additionally filters logs created by Microsoft libraries.

    • Logging to the console in CLI tool

    ...