AL Protocol Spec, Table 112 - IP Address, "IP Address may be accepted but should be valid"

[AlfredoAtSherpa]

Hello PI Forum colleagues:

I am at a loss in the interpretation of this clause. Since in the future compliance with this clause may be added to the ART as per post https://forum.profinet.com/viewtopic.ph ... 4f115f8ef5 (Is there a test case that enforces 4.3.1.4.21.2 Coding of the field IPAddress?), I want to make sure how to interpret this clause.
I have tested an ET200 part of our standard setup, and this device rejects DCP set of an IP address such as "192.168.0.0" will reject the value with a standard error response. This seems to be the better approach, as there is no problem with conformance to this clause with this behaviour.
But a device which we are developing, allows the setting of such value, i.e., does not issue error and reports back such IP setting to an DCP identify request.
My question is, what does the spec mean by "should be invalid"? Does it mean that if one sends a "ping" command the device does not reply?
Will be grateful for that answer, and more generally, is there any advice against allowing for the DCP settings of values such that Host part of the IPAddress is a series of consecutive "0", as the specification puts it?
Again, thanks a lot for the guidance and words of wisdom.

[Raik.Zachmann]

Hello PI Forum colleagues:

I am at a loss in the interpretation of this clause. Since in the future compliance with this clause may be added to the ART as per post https://forum.profinet.com/viewtopic.ph ... 4f115f8ef5 (Is there a test case that enforces 4.3.1.4.21.2 Coding of the field IPAddress?), I want to make sure how to interpret this clause.
I have tested an ET200 part of our standard setup, and this device rejects DCP set of an IP address such as "192.168.0.0" will reject the value with a standard error response. This seems to be the better approach, as there is no problem with conformance to this clause with this behaviour.
But a device which we are developing, allows the setting of such value, i.e., does not issue error and reports back such IP setting to an DCP identify request.
My question is, what does the spec mean by "should be invalid"? Does it mean that if one sends a "ping" command the device does not reply?
Will be grateful for that answer, and more generally, is there any advice against allowing for the DCP settings of values such that Host part of the IPAddress is a series of consecutive "0", as the specification puts it?
Again, thanks a lot for the guidance and words of wisdom.

Dear Alfredo,

thanks for bringing this up. We put a comment on these lines in the current PI-Review and asked for the expected behavior.

Best regards,
Raik

[AlfredoAtSherpa]

Dear Alfredo,

thanks for bringing this up. We put a comment on these lines in the current PI-Review and asked for the expected behavior.

Best regards,
Raik

Thank you so much for the follow-up, and for your patience with my inquires.