How to Add a Connector
Connectors are the objects that retrieve data from data sources so adding a Connector to a Core is a key step when starting to use system:inmation. To add a Connector you must first log in to DataStudio as an Administrator.
New Object Wizard
Once log in is complete, navigate to the I/O Model panel and expand the object tree to reveal the Core object. Connectors must always lie beneath a Core object in the object tree so the Connector is created by opening the context menu of the Core, selecting Admin > New > Services > Connector.
This will open the Create Object wizard.
On the left hand side of the wizard are the various configuration steps available for a fully detailed configuration. The right hand side the properties that can be configured per step. To only show the mandatory fields that need to be configured, click the Mandatory Properties only checkbox. These fields are indicated in the main wizard window by an asterisk. To navigate through the steps use the Next and Back buttons and click Create to send the creation command to the server.
In this example we will create a local Connector, meaning on that is on the same host as the Core Server.
-
In the first step of the Create Connector wizard, enter name and optionally a description like shown below. The other fields can be left blank so click Next.
| The name will be the last part of the object path; and therefore, is used whenever any interface wants to read from the object or its children. |
-
In the communication settings for the Connector, the Host Name or IP must be entered. In this case, as we are creating a local connector object we enter "localhost" in this field. Click Next to continue.
| The Communication settings are of special importance when your Connector is running on a remote location and listening on a different TCP port to the default one. Make sure you know these details when configuring the Connector. The Timeout property is crucial when long distance connections with high network lag times need to be used. If a TCP packet between Connector and Core needs more time than the specified timeout, the Core disconnects the Connector and initiates a reconnect. To avoid unnecessary reconnects over slow network connections, it is recommended to increase the timeout to a higher value. |
-
If you intend to use X.509-certificate-based authentication for Intercomponent Communication, select TLS_X509 for the Security Mode property and in the Self Certificate Selection subsection set Subject Name to the subject name of the certificate used by the connector service.
Otherwise the default Secure Remote Password (SRP) authentication will be used.
| For services using the X.509 security mode the corresponding objects need to be configured before the Service Installation. |
-
In the DCOM credentials options page any special account details can be entered if the Connector needs permission to read from a certain data source. In this example we can leave the fields empty and click Next.
-
In the Connector options page it can be set that the Connector will automatically enable any data sources as default and optionally bypass any discovery rules in place. There is also the option to disable the discovery mechanisms of certain OPC data sources.
The use of whitelists/blacklists to control datasource discovery is explained below. However, for this example we will leave the options as default, so click Next to continue.
-
The OPC UA Stack page allows for specific configuration of the connection to OPC UA data sources. This isn’t necessary for this example so click Next.
-
The Script Library allows you to set the script library compound that contains Lua-Scripts libraries.
-
In the Location properties page the specific location details of the Connector can be configured and also whether the location should be tracked. This isn’t necessary for this example so click Next.
-
In the Store and Forward Retention Method page, the buffering functionality of the Connector can be configured. This starts when the communication between Core and Connector is interrupted. In this case the Connector buffer all data it is supposed to pass on to the Core (including history data for objects whose value is archived). This is explained in detail in the system documentation. In this example we can leave the default values. This keeps buffering data on the local hard disk until a size of 20 GB is reached. Once the maximum size reached the new data coming in is kept, "old" data is purged from the buffer. Click Next to continue.
| Data purged from the Store and Forward buffer is lost. Make sure the specified GB size for the buffer is actually available on the hard disk that hosts the system:inmation installation. If memory is not there, the Data Source object will be in a warning (yellow) state in the model panel. |
-
In the following steps, further properties concerning Store and Forward, Transmission Limits and other custom properties can be configured. However, for this example these options can be skipped and we can click Next till the Summary page is reached and then click Create OR click Create on the current step to send the command to the server.
The new Connector should be visible in the I/O Model panel under the Core object. If there are any OPC servers available to the Connector they will be automatically visible in the tree below the Connector as Data Source items.
| Multiple Connectors instances can be created (and connected to the Core) on the same host. Please see the documentation on Installing multiple instances of services for more details. |
Datasource Discovery Whitelist/Blacklist
In the Connector Options property group, the discovery of Datasources by the Connector service can be configured via the Whitelist and Blacklist tables.
-
Click on the table icon to open the Datasource Whitelist table. In the table you can enter the full Object Name of the Datasource you wish to Whitelist (meaning, that the Datasource will be automatically added on discovery by the Connector service). You can also drag and drop a Datasource from the I/O model into the table.
-
Click Ok to save the Datasource to the Whitelist and don’t forget to click Apply in the Object Properties panel to confirm the changes.
| If the Whitelist is left empty then all datasources (except those on the Blacklist) will be automatically discovered and added. |
-
To add a Datasource to the Blacklist, click on the table icon to open the Blacklist table.
-
By default, the inmation OPC Classic server Datasource is on the Datasource blacklist. This means that the server will not automatically appear in the I/O Model when discovered by the Connector service. If the inmation server is removed from the Blacklist it will be discovered and then can be browsed in the I/O Model (this is not particularly recommended unless connecting to a server that is part of another inmation system).
-
Drag and drop another OPC datasource into the blacklist table (for example, the Matrikon simulation server) and click Ok. Then click Apply in the Object Properties panel to confirm the changes.
-
Delete the Matrikon Simulation Server Datasource in the I/O Model then restart the Connector service in Windows Task Manager.
-
After restart, the Datasources on the Blacklist are no longer added to the I/O Model underneath the Connector object.
| The Blacklist has priority over the Whitelist so if the same Datasource is added to both lists, the Blacklist overrules and the Datasource will not be automatically added to the I/O Model |
Under the Datasource Discovery Suppression property group it is possible to completely disable the automatic discovery of all OPC Classic Datasources or all OPC UA Datasources. By default, automatic discovery of OPC UA servers is disabled.