Loading ...
Sorry, an error occurred while loading the content.

Add. AI PPR2 Draft 9 has been uploaded

Expand Messages
  • Coleman Brumley
    IP-WG, Addendum AI PPR2 Draft 10 has been uploaded to the Yahoo site. Please disregard Draft 9, as I uncovered some things that were missed when I was
    Message 1 of 4 , Jan 16, 2012
    • 0 Attachment

      IP-WG,

       

      Addendum AI PPR2 Draft 10 has been uploaded to the Yahoo site.  Please disregard Draft 9, as I uncovered some things that were missed when I was preparing this notice.

       

      The  direct link is:

      http://groups.yahoo.com/group/bacnet-ip-wg/files/Add-135-2010ai-PPR2-Draft10.doc

       

      Draft 9 contains the following changes:

      > discuss extensibility in the Network_Type property value table

      > modified language in Network_Number writability description

      > discuss extensibility in the Command property value table

      > new language for Command property values regarding errors

      > new language for Command property values regarding Reliability

      > scrutinized language regarding "this device" vs. "this object" vs. "this port" for Command property value descriptions

      > separated IDLE from the Command property value table, as it's not a writable value

      > Added RESTART_AUTONEGOTIATION

      > "Data_Link_Type" has been changed to "Network_Type" throughout the document

      > The description of Network_Number has been modified to indicate that a PTP link has a value of 0 and is read-only.

      > Footnote 1 has been created and applied to Network_Number and footnotes have been renumbered.

      > Added blanket statement regarding persistence of values.  Removed this language in the property descriptions.

      > Made the routing table value member descriptions into a table.

      > Re-worded BACnet_IP_Mode value descriptions. 

      > used and explicit length when describing the 0x00 version of BBMD_Global_IP_Address

      > modified language about FD_ and BBMD_ properties and when the values should be ignored

       

      Outstanding items:

      > I would welcome feedback on the language for the RESET_PORT Command value.  What does it mean to reset a port?  Can the reset fail?

       

      Regards,

      Coleman

       

       

    • Isler, Bernhard
      Coleman, The following has been seen while looking through rev 10: The new Reliability enumerations need to be introduced by adding them to the clause 12
      Message 2 of 4 , Jan 17, 2012
      • 0 Attachment

        Coleman,

         

        The following has been seen while looking through rev 10:

         

        The new Reliability enumerations need to be introduced by adding them to the clause 12 preamble, on page 146.

         

        A possible text for the RESET_PORT (which I would prefer to rename to RESTART_PORT) could be:

         

        RESTART_PORT

        This port will restart and reconnect to the network as if the device were reinitialized. All data that was learned, cached or otherwise automatically determined for the port's operation shall be cleared. All initialization, negotiation and registration functions the port is expected to perform on device initialization shall be performed again.

        If the value of Changes_Pending is TRUE, writing this value shall result in an error response with ‘Error Class’ of PROPERTY and an ‘Error Code’ of VALUE_OUT_OF_RANGE.

        If the restart fails, the value of the Reliability property shall be RESTART_FAILURE and the value of this property shall be IDLE.

        In the exsiting languge for Nmax-master (Change to Clause 9..5.3), there is a reference to the Device object (first line), but this language should reference the respective NPO. Maybe clause 9 should be scanned for other such references to the Device object that may need to be adapted.

         

        Best,

        Bernhard

         

        Bernhard Isler
        Siemens Switzerland Ltd
        Building Technologies Division

        From: bacnet-ip-wg@yahoogroups.com [mailto:bacnet-ip-wg@yahoogroups.com] On Behalf Of Coleman Brumley
        Sent: Tuesday, January 17, 2012 3:07 AM
        To: bacnet-ip-wg@yahoogroups.com
        Subject: [bacnet-ip-wg] Add. AI PPR2 Draft 9 has been uploaded

         

         

        IP-WG,

         

        Addendum AI PPR2 Draft 10 has been uploaded to the Yahoo site.  Please disregard Draft 9, as I uncovered some things that were missed when I was preparing this notice.

         

        The  direct link is:

        http://groups.yahoo.com/group/bacnet-ip-wg/files/Add-135-2010ai-PPR2-Draft10.doc

         

        Draft 9 contains the following changes:

        > discuss extensibility in the Network_Type property value table

        > modified language in Network_Number writability description

        > discuss extensibility in the Command propert y value table

        > new language for Command property values regarding errors

        > new language for Command property values regarding Reliability

        > scrutinized language regarding "this device" vs. "this object" vs. "this port" for Command property value descriptions

        > separated IDLE from the Command property value table, as it's not a writable value

        > Added RESTART_AUTONEGOTIATION

        > "Data_Link_Type" has been changed to "Network_Type" throughout the document

        > The description of Network_Number has been modified to indicate that a PTP link has a value of 0 and is read-only.

        > Footnote 1 has been created and applied to Network_Number and footnotes have been renumbered .

        > Added blanket statement regarding persistence of values.  Removed this language in the property descriptions.

        > Made the routing table value member descriptions into a table.

        > Re-worded BACnet_IP_Mode value descriptions. 

        > used and explicit length when describing the 0x00 version of BBMD_Global_IP_Address

        > modified language about FD_ and BBMD_ properties and when the values should be ignored

         

        Outstanding items:

        > I would welcome feedback on the language for the RESET_PORT Command value.  What does it mean to reset a port?  Can the reset fail?

         

        Regards,

        Coleman

         

         


      • Hartman, John
        I just got called home for an injured dog, so I will have to miss the phone conference. Some comments. If any of these don t make sense, let me know, and I
        Message 3 of 4 , Jan 18, 2012
        • 0 Attachment

          I just got called home for an injured dog, so I will have to miss the phone conference.

           

          Some comments.  If any of these don’t make sense, let me know, and I will try to explain them by e-mail later today (or at least before Chicago).

           

          In 12.X Network Port Object

          Replace

           

          It is a local matter whether [CLB1] a port object contains properties specific to that port type. For example, if a network port object contains MS/TP properties, the device doesn’t necessarily have an MS/TP port.

          with

          As specified in Table 12-X and the text below, some properties of the Network Port Object are required if the object is used to represent a network of a given type.  For example, a Network Port Object whose Network_Type  is MSTP must include the Max_Master property, and a Network Port Object whose Network_Type is BACnet_IPV4 must include the BACnet_IP_Subnet_Mask property.  Aside from the properties so required, it is a local matter whether a Network Port Object contains properties that do not apply to its Network_Type.  For example, a Network Port Object whose Network_Type  is MSTP may include the BACnet_IP_Subnet_Mask property, although the value of this property would not be used by the network.  Some vendors may find it convenient to have all of their Network Port Objects support the same list of properties regardless of Network_Type.  This is permitted, but not required.

          The existing text talks only about odd cases that are ALLOWED, rather than what is EXPECTED.  My update attempts to explain WHY we allow this.

           

           

           

           

          1.       Table 12-X says Network_Number required writable, but 12.X.9 says required writeable only in routers.

           

          2.       12.X.9 says "...force the Network_Number_Quality to ..."  But Network_Number_Quality is OPTIONAL.  Do we need to add "if present"?"...force the Network_Number_Quality (if present) to ..."

           

          3.       Why is Default Gatewy required if DHCP?  Should it be superscript 7 rather than 8?

           

          4.       Table 12-X says BBMD_NAT_Traversal is required writable for BBMD.  Also Table 12-X says BBMD_Global_IP_Address is required, but it not required writeable.  The subclauses do not specify any required presence. 

            

          5.       FD_BBMD_IP_Address and FD_Subscription_Lifetime are NOT required writable? 

           

          6.       Command is required writable - MUST we allow editing the port object in all cases?

           

          7.       12.X.12 "The device will activate the currently visible configuration...".  Change to "The device SHALL activate...".  Similar for the remaining items, which already have some "shall" text.

           

          8.       Paragraph after the list of commands:  "When this property has a value of IDLE, then any pending commands have BEEN applied…”

           

          9.       "When making configuration changes to the CERTAIN PROPERTIES OF THE Network Port Object..."

             

          10.   12.X.21 requires multicast address to be writable, but Table 12-X does not   

              

          11.   12.X.31 FD_BBMD_IP_Address: Why include a UDP port?  Can't register foreign with a network on a different UDP port - wouldn't see any non-broadcast traffic from the network.   If it IS a HostNPort, then change the name to FD_BBMD_Address.   The description says "indicates the IP address", which doesn't really describe the HOSTNAME choice.

             

          12.   12.X.33 and .34: why is Max_Master required writeable, but Max_Info_Frames not  Table 12-X does not require writable   

              

          13.   12.X.37 talks about seeing if slaves support Who-Is.  I see that this is moved from the Device object.  Did I miss something?  How can a slave send an I-Am?

              

          14.   12.X.40 Routing Table "...contains the table of first hop routers to remote networkS REACHED THROUGH this port." (making “network” plural, and making clear which routers are included)

             

           

           

           

           

           


           [CLB1]0003-001 language




          The information contained in this message is privileged and intended only for the recipients named. If the reader is not a representative of the intended recipient, any review, dissemination or copying of this message or the information it contains is prohibited. If you have received this message in error, please immediately notify the sender, and delete the original message and attachments.
        • Hartman, John
          Sorry to miss the fun-fest this morning. Our elderly dog fell down the stairs. Much howling by dog and people. Sprained leg, nothing broken. My apologies if
          Message 4 of 4 , Jan 18, 2012
          • 0 Attachment

            Sorry to miss the fun-fest this morning.  Our elderly dog fell down the stairs.  Much howling by dog and people.  Sprained leg, nothing broken.

             

            My apologies if any of what follows has already been dealt with:

             

            Regarding BBMD_NAT_Traversal:  Table 12-X says it is required writeable for a BBMD.  12.X.26 says is must be writeable if present, but doesn’t require it to be present (mentioned in my previous e-mail)

            A more important question is: what is the intent?  Requiring this to be writeable would seem to require that all BBMD support NAT.  I don’t see any such requirement in Annex J, K, or L, or in the proposed changes to those annexes by this addendum.  I checked 135.2-2009, and the BTL Test Documents V9 and the letters “NAT” don’t even appear there.  If the intent is to require NAT support in all BBMD, we need a requirement to that effect elsewhere in the addendum.

             

            In Table 12-X, Slave_Proxy_Enable and Manual_Slave_Address_Binding are required writeable for MS/TP devices that support slave-proxy operation.  But 12.X.35 does not require Slave_Proxy_Enable to be writeable.

             

            As mentioned in my previous e-mail, 12.X.37 talks about seeing if slaves support Who-Is.  I see that this is moved from the Device object.  Did I miss something?  How can a slave send an I-Am?  Isn’t the whole POINT of Slave-Proxy that they CAN’T?

            The text for 22.1.5 item (c) clearly thinks slaves cannot send I-Have or I-Am.

            Or perhaps the intent is to include MASTERS that don’t support Who-Is?  If so, the properties need to be renamed.

             

            In 12.X.40 Routing_Table, the description for mac-address is

            “The MAC address of the next router on the path to that network number”

            I think this would be clearer as

            “The MAC address of the router on the network connected to this port that leads directly or indirectly to that network number”

             

            In J.4.3 BBMD Concept: just repeat the sentence added to J.5.2

            If a B/IP device is unable to be configured as a BBMD, then it shall be capable of being configured to register itself with a BBMD as a foreign device

             

            In J.5.2, the line-spacing is wider than in other clauses

             

            It’s probably just me, but throughout the addendum, the words “port” and “network” are used interchangeably.  Is this desirable?  Clause 6 calls these things “data links” most places, and “port” when talking about routers.  “Port”, of course has a very different meaning when talking about BACnet/IP and UDP ports.

            For example 12.X.12 says “…UDP port used by this port”.  The two distinct meanings of “port” seem awkward in the same sentence.  12.X.27 uses “UDP port on  which the network port can be reached”, which seems clearer.

             

            We DID agree to change Port_Type to Network_Type etc.

            Is there clarity or precision to be gained by using the same word (either “port” or “network”) or phrase (“network port”) everywhere?

             

            And if you want to tell me “Hartman, get a new hobby”, I can live with that too (but it will still bother me)

             

            Browsing through the standard, it seems that the usual style is to capitalize the word “object” in titles, and not to capitalize it when referred to in the text.  Thus my proposed text in 12.X should be “some properties of the Network Port object”, while the title of 12.X remains “Network Port Object”.  The capitalized “Object” appears several times in the proposed text where lower case should be used.

             

            In 12.11.18 Max_APDU_Length_Accepted, change

            “shall be less than or equal to the largest APDU_Length of the enabled Network Port Object used to represent the underlying data link technology.” to

            “shall be less than or equal to the largest APDU_Length of the enabled Network Port Objects used to represent the underlying data links.”

             

            Make it clear that there may be MULTIPLE Network Port objects.  “Technology” makes sense in the clause about “…constrained by…”, but not when referring to the NP object.  I might argue about “data link”, however J

             

            In 12.11.32 Max_Master, and 12.22.33 Max_Info_Frames, need text to describe the value if there are multiple MS/TP datalinks.  Personally, I would REMOVE footnote “X” from Table 12-13 and move that text to 12.22.32 and 12.11.33 where it is more visible.  The footnotes in the object tables should apply to the Conformance Code column, and footnote X does not.

            The text under 12.11.32 says “…on the MS/TP network connected to this port”.  What “port” are we talking about?  This is the Device object.

             

            I propose this text for 12.11.32:

            The Max_Master property, of type Unsigned , shall be present and writable if the device is a master node on an MS/TP network. The value of Max_Master specifies the highest possible address for master nodes on the MS/TP network and shall be less than or equal to 127. See 9.5.3.

             

            If the device supports multiple MS/TP networks, then this property shall reflect the value of the Max_Master property of the enabled Network Port object with the lowest object instance whose Network_Type is MSTP.

             

            And for 12.11.33:

            The Max_Info_Frames property, of type Unsigned, shall be present if the device is a node on an MS/TP network. The value of Max_Info_Frames specifies the maximum number of information frames the node may send before it must pass the token.  This property shall have a minimum value of 1 and a maximum value of 255.  If Max_Info_Frames is not writable or otherwise user configurable, its value shall be one. See Clause 9.5.3.

             

            If the device supports multiple MS/TP networks, then this property shall reflect the value of the Max_Info_Frames property  of the enabled Network Port object with the lowest object instance whose Network_Type is MSTP.

             

            In 15.5.2, the existing text says “contains the value 'Device Object' and…”, while the new text about ‘Network Port’ omits the word “object”.  I believe that both should use “object” uncapitalized.

             

            In the ASN.1 production for BACnetBDTEntry, does it make sense to use BACnetHostNPort?  I can’t see how a BBMD can work if all the entries in the BDT don’t have the same UDP port: the same value as BVLL uses for non-broadcast traffic.  Otherwise, the remote BBMD would need to listen on BOTH the port listed in the BDTs, to see forwarded broadcasts, AND on the “real” port, in order to see local broadcasts and unicasts.  I know that the BVLL message includes the port, but I think it is silly there as well.  This probably affects several properties in the Network Port, but I will limit to comment to this one place, for your convenience in telling me to go away.

             

            In the ASN.1 for BACnetHostAddress, “octet octet” seems silly.  How about “0, 4, or 16 octets”?  The OCTETSTRING datatype is already specified.

            When would this be zero octets?  If this means “not used”, better to add a NULL CHOICE to BACnetHostAddress.  Elsewhere in the standard a zero-length address means BROADCAST, which is clearly NOT the meaning here.

             

            In 22.1.5 Minimum Device Requirements, a more logical order would group the objects and services together, rather than the current order-we-added-them-in:

            (a) contain exactly one Device object,

            (b) contain a Network Port object for each configured network port.

            (c)  execute the Who-Has and Who-Is services (and thus initiate the I-Have and I-Am services) unless the device is an MS/TP slave device,

            (d) execute the ReadProperty service,

            (e) execute the WriteProperty service if the device contains any objects with properties that are required to be writable, and

            (f) execute the WriteProperty service if the device executes the WritePropertyMultiple, AddListElement or RemoveListElement services, and

            (g) allow the WriteProperty service to modify any properties that are modifiable by the AddListElement or RemoveListElement services

             

            (I might also point out that requiring WriteProperty to modify anything that can be manipulated with Add/Remove list is a problem for long lists on devices without segmentation.  But that is for another day)

             

             

             





            The information contained in this message is privileged and intended only for the recipients named. If the reader is not a representative of the intended recipient, any review, dissemination or copying of this message or the information it contains is prohibited. If you have received this message in error, please immediately notify the sender, and delete the original message and attachments.
          Your message has been successfully submitted and would be delivered to recipients shortly.