Publish Response with Bad_TooManyPublishRequests

[cacamille3]

The Publish mechanism works well between my server and your client.
But if I send a Bad_TooManyPublishRequest PublishResponse, after my Server PublishRequest queue had been fulled, it is not working anymore.

Should I sent a special sequenceNumber ? (I tried normal +1 of last acknowledged and also 1), a couple of StatusCode and ServiceResult ? Or should I really put a bigger queue (even if it do not resolve the problem...) and hope I will have some NotificationData to send before it will be fulled...

Regards.

CG

[cacamille3]

If I put the SequenceNumber for the Bad_TooManyPublishRequests PublishResponse as the next one that will be used by a "normal" PublishResponse the Server do not acknowledge this next normal Publish Response but will acknowledge the next one (so +2 than the last acknowledge) and then ask for Republish for the +1 it didnt acknowledge (even if I sent it...)

So is it a normal behaviour or just a special case of UAExpert. There is not so much information about Bad_TooManyPublishRequests PublishResponse in the specification...

[Support Team]

We can't see a problem in the UaExpert, our tests worked fine. We have tested against our four (4) DemoServers in ANSI C, CPP, .NET and Java and we have tested against the .NET Reference Server of the OPC Foundation.

At this stage of the investigation we think that UaExpert behaves correct and also our SDKs behave correct. The behavior of Publish and Republish, including the acknowledgements and error scenarios are described in detail in the latest spec.

[Support Team]

This topic is solved and has been locked by the administrator.