Log In
Home
07/06/2017
Offline
Hello,
We have several questions related to the concept of AggregatingServers and Endpoints:
1/ We found the concept DataItem which describes links to automation dat, i.e. data not stored by a server but accessible by it (Part 8, §4). In the case of an AggregatingServer, where data are stored in subservers, but are not automation data, are the nodes necessarily typed by a subtype of DataItemType?
2/ If not, how does a Client know if the data it tries to fetch are “local to the server” or “available only on subservers”?
3/ In other terms, is the concept of aggregation covered by the OPC UA Specification, or is it a server-specific behavior?
4/ Is the endpointUrl field or serverUri field of the CreateSessionRequest (Part 4, §5.6.2) related to the concept of aggregation?
Thanks,
Best regards
In general aggregation can be transparent – i.e. the client does not see and know and should not care where the information comes from. Such an implementation is server-specific.
I assume that your questions relate to non-transparent aggregation. OPC UA provides some means for this:
* The ServerArray represents an array of ServerURIs that this Server makes use of (defined in Part 5).
* References (returned by the Browse Service) may reference nodes in such a remote server. If a Client finds such a remote reference it has to connect to the remote Server to access the referenced remote node.
* If the remote Server can only be accessed through the aggregating server, there has to be gateway information available in the discovery services (see UA Part 4). The Client will then connect to the aggregating server and specify the sub-server URI in the serverUri field of the CreateSession service.
1 Guest(s)
Usage Policy