Whether any specific approach is better or worse depends on the details.

i.e. it is possible to design a custom NodeManager that ends up using more RAM than standard class based model if simultaneous access is needed from many clients. You will need to benchmark different design approaches to determine which best meets your requirements.

Another question after I have started coding the static parts where it is ok to cache nodes in PredefinedNodes:

I have modelled an object type with a number of properties and variables in a ModelDesign.xml file. Then I create an instance of that type at server startup. Some of the properties and values are dynamic and their values are read from a controller. In order to do that I have registered a read eventhandler on their OnReadValue (BaseVariableState::OnReadValue) event. The read operation to the controller, which is invoked in the eventhandler, is async, but I wait for it to return in the handler (blocking). The MemoryBuffer example rely on the OnReadValue event to handle reads...

Will this solution lead to bad performance? Is it better to override CustomNodeManager2's virtual read method which has a list of nodes to read as input. Here it will be necessary to "route a node to its corresponding read handler", but it would be possible to make several async read operations to the controller for multiple nodes at the same time and wait for all before returning. In this way there would be several async operations running in parallel.

Or is the only right way to create an instance on demand and fill in its current values and never cache even smaller static parts.

If you need to optimize memory/lookup you need a custom nodemanager and hand code most of the operations.

However, you can instantiate the auto-generated classes as part of processing request.

i.e. create a instance of a generated class, fill it its current values an pass it to other code that uses the standard API to read the values.

we would need to check the details of the certification as well as the behavior to see whether there was a gap in testing which caused this to be missed or whether this is actually compliant behavior under the given circumstances.

Since we would not like to discuss this in public, can you please send a mail to compliance@opcfoundation.org with further details about the product, the version you are using and in which scenario you are experiencing the issue?

We will then investigate and get back to you via e-mail. If we find something missing in the testing we will then create additional test cases to cover those cases and make sure these are caught in the future.

Regards,
Alexander Allmendinger

I have an idea to model static parts (parts that are the same across all controllers) of a controller in a ModelDesign.xml file to utlize the framework as much as possible. As I understand it you get a lot out of the box (such as browsing) and support by the partial generated classes where you can put your implementation code. The static parts are not too extensive and can be cached in PredefinedNodes or similar.

Then treat the dynamic parts (parts that differ from controller to controller) more like the MemoryBuffer with no or sparse caching and custom browsing.  The question is how I can best achieve this? Same or different nodemanager for the two parts? Which nodemanager to use/start off from? Represent dynamic parts with NodeState-derived classes (like MemoryTagState) or custom lightweight classes? Are there cases where it is a good idea to skip using NodeState-derived classes?

I'm getting a bit confused by all different examples, and I find it difficult to become any wiser by reading the documentation. I can't find a user-friendly documentation which can guide me in the correct direction. There are som many different examples starting off from different nodemanager classes or interfaces (CustomNodeManger2, SampleNodeManager, INodeManager, INodeManager2...). It is difficult to get a clear view of the differences between the different nodemanagers. Seems like a tedious task to implement a nodemanager interface from scratch. Dynamic data is sometimes represented by NodeState and somtimes by custom lightweight classes. And so on...

I would really appreciate some help!    

Disclosure first: I am the author of node-opcua and I work at Sterfive, which makes one of the tools below.

1. The OPC Foundation now has its own editor. The UA NodeSet Editor (https://uanodeseteditor.opcfou.....ation.org/) is a free web tool to create and edit information models, start from the UA Cloud Library, and export a NodeSet. It can also check a Companion Specification Word document against its NodeSet. Source: https://github.com/OPCFoundati.....eSetEditor

2. UA-ModelCompiler is still the way to go if you need the generated C# or ANSI C classes the original question was about. Stefan's tutorial above remains a good introduction.

3. If what you need is the NodeSet2.xml itself, another approach is to write the model as text and compile it. That is what we built: OPC UA Modeler takes a small YAML description and generates a standard NodeSet2.xml, with the companion specifications (DI, Machinery, Robotics…) available by alias and the model validated as you type. Because the model is a text file, it diffs, reviews and merges like source code, and can be built in CI.

namespaceUri: http://acme.example/UA/Pump/
version: 1.0.0

namespaces:
- di

objectTypes:
- browseName: PumpType
  subtypeOf: di:DeviceType
  components:
  - browseName: di:ParameterSet
    components:
    - browseName: FlowRate
      dataType: ua:Double

To be clear about its limits: it produces NodeSet2.xml, not ModelDesign XML, and it does not generate C# classes. The .NET Standard stack, open62541, node-opcua and the other SDKs load the resulting NodeSet2.xml directly.

The language specification (CC BY 4.0), its JSON Schema and example models are public: https://github.com/node-opcua/.....ua-modeler
A tool-neutral walkthrough of what a NodeSet2.xml contains and the different ways to produce one: https://opcua-modeler.doc.ster.....deset2-xml

Etienne

A follow up question on ConditionName.

We have observed OPC UA servers that provide an empty or null ConditionName, while at the same time holding OPC Foundation certification. 

Given that the confirmation that ConditionName must not be null or empty, we are trying to understand how such a server could still pass conformance testing. Is ConditionName perhaps covered by a different Facet or Profile than those included in the server's certification (see below)? Or is the value simply not validated as part of the relevant conformance test cases?

It would be useful to understand whether this behavior is outside the scope of the certification tests, whether there is an interpretation issue regarding the specification requirements, or whether there is a gap in the current test coverage.

Any clarification on how ConditionName is evaluated during OPC Foundation certification would be greatly appreciated.

The A&C facet that the OPC UA servers holding OPC Foundation certification on are:

- A&C Base Condition Server Facet
- A&C Acknowledgeable Alarm Server Facet

Thanks

ConditionName identifies the Condition instance that the Event originated from. It can be used together with the SourceName in a user display to distinguish between different Condition instances. If a ConditionSource has only one instance of a ConditionType, and the Server has no instance name, the Server shall supply the name element of the BrowseName of the ConditionType.

Specification says no and provides a default if it does not make sense.

Another question in same area. 

Is it allow to set property ConditionName (in EventType ConditionType) to null or empty according to the OPC UA specifications. 

Thanks

OPC 10000-24, OPC Unified Architecture - Part 24: Scheduler
http://www.opcfoundation.org/UA/Part24/
OPC 10000-25, OPC Unified Architecture - Part 24: Scheduler
http://www.opcfoundation.org/UA/Part24/
OPC 10000-26, OPC Unified Architecture - Part 24: Scheduler
http://www.opcfoundation.org/UA/Part24/

For "OPC 10000-24, OPC Unified Architecture" this is fine, but not for "OPC 10000-25..." and "OPC 10000-26..."

This need to be 

OPC 10000-24, OPC Unified Architecture - Part 24: Scheduler
http://www.opcfoundation.org/UA/Part24/
OPC 10000-25, OPC Unified Architecture - Part 25: ObjectSerialization
http://www.opcfoundation.org/UA/Part25/
OPC 10000-26, OPC Unified Architecture - Part 26: LogObject
http://www.opcfoundation.org/UA/Part26/

Btw: the underlying links for OPC 10000-25 and -26 are correct

And the Bibliography is only in the PDF, not downloaded markdown or browsed spec.

But the Value cannot be Variant.Empty for mandatory properties.

However, the "" rule does not apply if the description of the property makes it clear that "" is not a valid value:

SourceName provides a description of the source of the Event. This could be the string-part of the DisplayName of the Event source using the default locale of the server, if the Event is specific to a Node, or some server-specific notation.

(i.e. "" is not a description)

Clients should be designed defensively to deal with misbehaving Servers but that does not make the server right.

Is it allowed, according to the specifications, to set the value on a mandatory property, in an event, to null. 

For example:

Assuming an event of EventType BaseEventType.

Is it allowed for an OPC UA server to set, for example, property SourceName to null in the event?

If it is, is it the same for any mandatory property in any EventType? 

Should an OPC UA client be able to take that into account?

Thanks