Diversity Agents

Replication

The chapter below refers to DiversityCollection but is valid for DiversityAgents as well.

Replication

If you wish to work with your data on a local database (called subscriber), e.g. on your laptop, not linked to a database on a central server (called publisher) and these data should be synchronized with the data in the database on the server, you may use the replication function of DiversityCollection. To install the database on your local computer see the installation section.

To use the replication function you require the roles Replicator or Administrator.

Add Publisher

To define a publishing database choose Data → Replication → Add Publisher from the menu. A window will open where you choose the publisher. After the publisher is set, you may transfer data between your local database (subscriber) and the publisher. This function is only available for administrators.

Remove Publisher

To remove a publisher from the list choose Data → Replication → [Publisher] → Remove from the menu (where [Publisher] is the name of the publishing database on the publishing server). This function is only available for administrators.

Clean database

Initially you may wish to remove all previous data from your local database (subscriber). Choose Data → Replication → Clean database … from the menu. A window will open as shown below where you may choose the ranges which should be cleared:

  • Definitions = the basic definitions within the database, e.g. the available taxonomic groups.
  • Descriptions = the descriptions and their translations of the tables and columns of the database.
  • Project, User = the available projects and users.
  • Basic data = basic data like the collection.
  • Data = the specimen, organisms etc.

Choose the data ranges you wish to clear and click on the button. All tables which contain data will be listed as shown below.

 

Choose the tables which should be cleared and click on the Clean database button. Please keep in mind that you can not delete data from a table as long as there is data in a related table depending on the data you wish to delete. The sequence of the tables is organized to avoid these problems.

Download

To download data from the publisher choose Data → Replication → [Publisher] → Download from the menu (where [Publisher] is the name of the publishing database on the publishing server). A form will open as shown below. Choose the project of the data and the data ranges (see above) which you wish to download. Click on the button to list the tables containing data. To start the download click on the Start download button. With the Force download, ignore conflicts option you can decide whether or not the data in your local database (= Subscriber) should be checked for changes before you download the data from the publisher.

If not all data should be included in the replication, you have the option to set a filter. Click on the button for the table where the data should be filtered to set this filter. A window as shown below will open.

All columns of the table will be listed and allow you to set the filter. To inspect the filtered data, click on the button. Click on the button to see the current filter. If a filter is set this will be indicated with a blue background .  

Merge

To merge data from your local subscriber database with the publisher you must first choose a project. Choose Data → Replication → [Publisher] → Merge** from the menu ([Publisher] is the name of the publishing database on the publishing server). As described for the download, choose the data ranges and click on the button. To start the upload click on the Start merge button.

 

Upload

To transfer data from your local subscriber database to the publisher you must first choose a project. Choose Data → Replication → [Publisher] → Upload from the menu ([Publisher] is the name of the publishing database on the publishing server). As described for the download, choose the data ranges and click on the

button. To start the upload click on the Start upload button. With the Force upload, ignore conflicts option you can decide whether or not the data in server (= Publisher) should be checked for changes before you upload the data from your local database (= Subscriber).

As described for the download, data may be filtered with a click on the button (see above).

 

Tools

To fix problems that may interfere with the replication you find some tools under the menu Data → Replication → [Publisher] → Tools… from the menu ([Publisher] is the name of the publishing database on the publishing server). A window will open as shown below.

You may synchronize the RowGUIDs between basic subscriber and publisher tables if for any reason these are differing, e.g. due to manual insert. Choose the table that should be synchronized. The tables will be compared for both publisher and subscriber. The datasets with identical key but different RowGUID will be listed (see above). Click on the Start update button to synchronize the RowGUIDs.  

Conflict

If the transfer of data was successful, the numbers of the transferred data will be shown as below.

During the download or upload a conflict may occur, if the data has been edited in both databases. This will be indicated as shown below.

Click on the button to open a window as shown below where you can choose between the two versions of the data as found in the publisher and the subscriber database.

The conflicting columns are marked red. For text values the program will create a combination of both values (see above) in a merged version of the data. Choose the preferred version of the data and click Solve conflict button. If you can not solve a conflict, use the Ignore conflict or Stop conflict resolution buttons respectively.

 

Report

At the end of each transfer a report will be created with a summary for every table which has been included.