rfc2127.txt

来自「RFC 的详细文档！」· 文本 代码 · 共 1,729 行 · 第 1/5 页

   ifInMulticastPkts Return zero.

   ifInBroadcastPkts Return the number of frames received
                     on this signaling channel with TEI=127.

   ifInDiscards      The total number of received frames which have been
                     discarded.
                     The possible reasons are: buffer shortage.

   ifInErrors        The number of inbound frames that contained
                     errors preventing them from being deliverable
                     to the signaling channel.

   ifInUnknownProtos Return zero.

   ifOutOctets       The total number of octets transmitted on this
                     signaling channel.

   ifOutUcastPkts    The number of frames transmitted on this
                     signaling channel whose address is not TEI=127.

   ifOutNUcastPkts   Deprecated.  Return the number of frames
                     transmitted on this signaling channel with TEI=127.

   ifOutMulticastPkts
                     Return zero.

   ifOutBroadcastPkts
                     Return the number of frames transmitted
                     on this signaling channel with TEI=127.






Roeck                      Standards Track                     [Page 13]

RFC 2127                        ISDN MIB                      March 1997


   ifOutDiscards     The total number of outbound frames which
                     were discarded. Possible reasons are:
                     buffer shortage.

   ifOutErrors       The number of frames which could not be
                     transmitted due to errors.

   ifOutQlen         Deprecated. Return zero.

   ifSpecific        Deprecated. Return {0 0}.

3.3.  Relationship to other MIBs

3.3.1.  Relationship to the DS1/E1 MIB

   Implementation of the DS1/E1 MIB [12] is not required for supporting
   this MIB. It is however recommended to implement the DS1/E1 MIB on
   entities supporting Primary Rate interfaces.

3.3.2.  Relationship to the DS0 and DS0Bundle MIBs

   Implementation of the DS0 MIB [13] is optional.

   Implementation of the DS0Bundle MIB [13] may be required only if
   hyperchannels are to be supported, depending on the multiplexing
   scheme used in a given implementation. See chapter 3.4.2 for details
   on how to implement hyperchannels.

3.3.3.  Relationship to the Dial Control MIB

   Implementation of the Dial Control MIB [15] is required.

3.4.  ISDN interface specific information and implementation hints

3.4.1.  ISDN leased lines

   ISDN leased lines can be specified on a per-B-channel basis.  To do
   so, the value of isdnBearerChannelType has to be set to leased(2).
   There is no signaling protocol support for leased line B channels,
   since there is no signaling protocol action for these kinds of
   interfaces.










Roeck                      Standards Track                     [Page 14]

RFC 2127                        ISDN MIB                      March 1997


   If there is no signaling support available for an ISDN interface,
   this must be specified in the appropriate interface specific table.
   For Basic Rate interfaces, isdnBasicRateSignalMode of
   isdnBasicRateTable must be set to inactive(2).  For Primary Rate
   interfaces, dsx1SignalMode of dsx1ConfigTable in DS1/E1 MIB [12] must
   be set to none(1).  There are no isdnLapdTable or isdnSignalingTable
   entries for such interfaces.

   Depending on the leased line type and the service provider, the D
   channel can be used for data transfer.  If this is the case the D
   channel interface type is ds0(81) instead of lapd(77) and its usage
   is identical to B channel usage if there is no signaling channel
   available.

   For a Primary Rate interface which is entirely used as a leased line,
   there is no ISDN specific information available or required.  Such
   leased lines can entirely be handled by the DS1/E1 MIB.

3.4.2.  Hyperchannels

   The active switch protocol defines if hyperchannels are supported,
   and the actual support is implementation dependent.  Hyperchannel
   connections will be requested by the interface user at call setup
   time, e.g. by the peer connection handling procedures.

   In the ISDN MIB, the isdnBearerMultirate object of isdnBearerTable
   can be used to check if hyperchannels are being used for an active
   call.

   If hyperchannels are being used, multiplexing between the
   encapsulation layer and the B channels is required, since there is
   one encapsulation layer interface connected to several B channel
   interfaces.  This can be accomplished in two ways.

   o    The DS0Bundle MIB [13] can be used to provide the multiplexing.
        See the DS0Bundle MIB document for details.

   o    The ifStackTable can be used to provide the multiplexing.  In
        this case, there are several ifStackTable entries with the same
        value of HigherLayer, and different values of LowerLayer.

   It is up to the implementor to decide which multiplexing scheme to
   use.

   Each hyperchannel call is treated as one call in the
   isdnSignalingStatsTable, independent of the number of B channels
   involved.




Roeck                      Standards Track                     [Page 15]

RFC 2127                        ISDN MIB                      March 1997


   For a hyperchannel call, all objects in the isdnBearerTable entries
   related to this call (i.e., all isdnBearerTable entries associated to
   B channels used by the hyperchannel) have identical values.  The
   related objects in the isdnBearerTable are:


       isdnBearerPeerAddress
       isdnBearerPeerSubAddress
       isdnBearerCallOrigin
       isdnBearerInfoType
       isdnBearerMultirate
       isdnBearerCallSetupTime
       isdnBearerCallConnectTime
       isdnBearerChargedUnits

3.4.3.  D channel backup and NFAS trunks

   D channel backup is defined in Q.931 [8], Annex F.  It describes Non-
   Associated signaling and its use and functionality is basically
   identical to Non Facility Associated Signaling (NFAS) trunks.

   Non Facility Accociated Signaling (NFAS) basically means that a D
   channel on a PRI interface is used to manage calls on other PRI
   trunks.  This is required in North America for H11 channels, since
   all 24 time slots are being used for B channels.

   According to Q.931, Annex F, the D channel backup feature can be
   provided on a subscription basis and is network dependent.  The D
   channel backup procedure is described in detail in Q.931.

   For D channel backup, the controlling isdnSignalingTable entry is
   layered on top of all attached LAPD interfaces.  This layering is
   done using the ifStack table.  There is only one active LAPD
   interface, however.  Inactive LAPD interfaces have an ifOperStatus of
   dormant(5).

   NFAS trunks are also handled using the ifStack table. In this case, a
   signaling channel is layered on top of a LAPD interface as well as on
   top of all physical interfaces which are controlled by the signaling
   channel, but do not supply a D channel.

3.4.4.  X.25 based packet-mode service in B and D channels

   X.25 based packet mode service over B channels can be handled using
   the Dial Control MIB by creating an appropriate peer entry.  The peer
   entry ifType can then be x25(5), thus providing access to X.25
   service.




Roeck                      Standards Track                     [Page 16]

RFC 2127                        ISDN MIB                      March 1997


   X.25 based packet mode service over D channels can be handled by
   creating an ifEndpointTable entry with an isdnEndpointIfType of
   x25ple(40).  The upper protocol layers can then be attached to this
   interface using the ifStack table.

3.4.5.  SPID handling

   Service Profile IDentifiers (SPIDs) are defined for BRI interfaces
   only, and being used in North America.  SPIDs are required for DMS-
   100, NI-1 and NI-2, and are optional for 5ESS.  A switch can define
   up to 8 SPIDs per BRI.

   Each Terminal Endpoint has a SPID assigned.  It is normally built
   from the party number (calling address for outgoing calls) with a
   number of digits prepended and appended.  Since each network appears
   to be different, both the calling address and the SPID have to be
   stored.

   The SPID identifies the particular services that have been
   provisioned for a terminal. If there are two B channels on a BRI,
   there can be two SPIDs, one for each of the two B channels.  There
   can also be a single SPID, providing access to both B channels.

   The SPID gets registered with the switch after link establishment.
   There is one data link for each SPID. As part of terminal
   registration, an EID (Endpoint IDentifier) is defined by the switch.
   On incoming calls, the switch may provide the EID, a called party
   number, or both, depending on the ISDN code implemented in the
   switch.

   The EID has two bytes: USID (User Service IDentifier) and TID
   (Terminal IDentifier). These are later used by some of the software
   versions running on the switch side (e.g. compliant with NI-1, 5ESS
   custom) to broadcast SETUP messages with these included, so the
   correct endpoint would accept the call. Other switch software
   versions identify the endpoint with the Called Party Number.

   In the ISDN MIB, the SPID can be entered using the isdnEndpointSpid
   object of isdnEndpointTable.  The isdnSignalingCallingAddress,
   already being used to specify the calling number, cannot be used to
   record the SPID since the values of the SPID and the Calling Address
   may differ and both may be required to be present.

3.4.6.  Closed User Groups

   Closed User Groups (CUG), as defined in I.255.1 [14], are supported
   for circuit mode calls by ETSI (ETS 300 138) and 1TR6.  In these
   networks, an ISDN address can have one or more Closed User Groups



Roeck                      Standards Track                     [Page 17]

RFC 2127                        ISDN MIB                      March 1997


   assigned.  If there is more than one Closed User Group assigned to a
   given address, one of those is the preferred Closed User Group.  For
   such addresses, only calls from assigned Closed User Groups are
   accepted by the network.

   Thus, Closed User Groups are a parameter for peer entries and are
   defined in the Dial Control MIB. A peer entry attached to a Closed
   User Group has to point to an ISDN interface which is attached to the
   Closed User Group in question.

3.4.7.  Provision of point-to-point line topology

   In the ISDN standards, there are two different meanings for the term
   "point-to-point".

   In ISDN standards, the term point-to-point are usually used for data
   link connections, i.e. layer 2 connections, where each layer 2
   connection from the TE to the network is a single point-to-point
   connection.  Multiple connections of this kind may exist on one
   physical (layer 1) connection, however, and in case of Basic Rate
   interfaces there may be several TE's connected to one physical line
   to the network.

   The second meaning of "point-to-point" refers to the line topology,
   i.e.  to layer 1 connections.  For Primary Rate interfaces, the line
   topology is always point-to-point.  For Basic Rate interfaces, layer
   1 point-to- point connections do exist in several countries, usually
   being used for connecting PBX systems to the network.

   The second meaning (layer 1 connections) is what will be referred to
   as "point-to-point" connection throughout this document.

   For Basic Rate interfaces, the isdnBasicRateTable object
   isdnBasicRateLineTopology can be used to select the line topology.

3.4.8.  Speech and audio bearer capability information elements

   The objects speech(2), audio31(6) and audio7(7), as being used in
   isdnBearerInfoType, refer to the Speech, 3.1 kHz Audio and old 7 kHz
   Audio (now Multi-use) bearer capabilities for ISDN, as defined in
   Q.931 [8], chapter 4.5.5, octet 3 of bearer capability information
   element.

   These capabilities are signaling artifices that allow networks to do
   certain things with the call.  It is up to the network to decide what
   to do.





Roeck                      Standards Track                     [Page 18]

RFC 2127                        ISDN MIB                      March 1997


   The Speech Bearer Capability means that speech is being carried over
   the channel, as in two people talking.  This would be POTS-type
   speech.  The network may compress this, encrypt it or whatever it
   wants with it as long as it delivers POTS quality speech to the other
   end.  In other words, a modem is not guaranteed to work over this
   connection.

   The 3.1 kHz Audio capability indicates that the network carries the
   3.1 kHz bandwidth across the network.  This would (theoretically)
   allow modem signals to be carried across the network.  In the US, the
   network automatically enters a capability of 3.1 kHz Audio on calls
   coming into the ISDN from a POTS network.  This capability restricts
   the network from interfering with the data channel in a way that
   would corrupt the 3.1 kHz VoiceBand data.

   7 kHz Audio was meant to signal the use of a higher quality audio
   connection (e.g., music from radio).  It was changed to Multi-Use
   capability to allow it to be used for video-conferencing with fall
   back to audio.

   In some cases, the Speech or 3.1 kHz Bearer Capability provides a 56
   kbit/s data path through the network.  Therefore, some people are
   setting up calls with the Speech or 3.1 kHz BC and transmitting 56
   kbit/s data over the connection.  This is usually to take advantage