Functional Specification For SNMP
Functional Specification For SNMP
(VMware)
Nokia is committed to diversity and inclusion. We are continuously reviewing our customer documentation and consulting with standards
bodies to ensure that terminology is inclusive and aligned with the industry. Our future customer documentation will be updated accordingly.
This document includes Nokia proprietary and confidential information, which may not be distributed or disclosed to any third parties without
the prior written consent of Nokia.
This document is intended for use by Nokia’s customers (“You”/”Your”) in connection with a product purchased or licensed from any company
within Nokia Group of Companies. Use this document as agreed. You agree to notify Nokia of any errors you may find in this document;
however, should you elect to use this document for any purpose(s) for which it is not intended, You understand and warrant that any
determinations You may make or actions You may take will be based upon Your independent judgment and analysis of the content of this
document.
Nokia reserves the right to make changes to this document without notice. At all times, the controlling version is the one available on Nokia’s
site.
This document is Nokia’ proprietary and confidential information, which may not be distributed or disclosed to any third parties without the
prior written consent of Nokia.
Copyright and trademark: Nokia is a registered trademark of Nokia Corporation. Other product names mentioned in this document may be
trademarks of their respective owners.
© 2023 Nokia.
Contents
1 Introduction...................................................................................................................................................... 5
1.1 General Principles..................................................................................................................................... 5
4 Feature summary...........................................................................................................................................35
4.1 NMS requests..........................................................................................................................................35
4.2 Spontaneous traps.................................................................................................................................. 36
8 MIB structure..................................................................................................................................................51
9 List of abbreviations..................................................................................................................................... 54
1 Introduction
This document describes from a functional point of view (that is, as seen by operators in charge of net-
work management) the SNMP-based northbound FM/CM-oriented interface (in the following called
NBI) between NetAct (as Nokia Element Manager, EMS) and Operations Support Systems (OSS) on
Network Management Layer.
While the lower management interfaces between network elements (NE) and NetAct are usually man-
ufacturer-dependent (because of various supported NE types and technologies with different manage-
ment interfaces and object models), the main reason for the provision of NBI is to facilitate an easy in-
tegration of Nokia (and related third-party) equipment in a multi-vendor environment and to provide a
unified network management view despite of the details of underlying interface technologies.
To fulfill these requirements the SNMP Agent provides a mediation application function (based on Net-
Act FM and CM platforms) which converts and maps the management information received by or gen-
erated in NetAct into the information required at NBI.
• Support fault management related supervision of network elements (and NetAct itself) used in
different domains and network technologies. To support fault management features, information
about the topology of the managed network (so-called CM) is needed as well.
• The interface is based on generic, domain & technology independent alarm (ITU-T X.733) and
topology data definitions. Despite of this generic approach, the operator working on NMS is able
to unambiguously identify the origin of an alarm (physical or logical faulty network resource) in the
managed network.
Note: In this document, the term SNMP Agent means the functionality provided by different
products of the NetAct Base family towards higher management systems, while the NMS
provides the SNMP Manager functionality. Both entities use the defined MIB for the provision
of the described northbound interface features.
Note:
To give to a superior NMS a real-time supervision capability about the managed network, NetAct has
to provide information about:
• Alarm and topology-related event occurrence in NetAct, NE and about connections between Net-
Act and NE.
• Other events, for example, notification informing about start and controlled shut-down of the SN-
MP Agent.
• Network topology (for example, which network elements and network resources the network is
composed of).
For this purpose, the SNMP Agent forwards spontaneous notifications (traps) in real-time to the regis-
tered NMS to indicate, for example:
• The occurrence, acknowledgment, changing and clearing of alarms within NetAct or NE, so that
related alarm information can be updated.
• Creation or deletion of network resources, so that data about network topology can be updated.
Each connected NMS is able to control the real-time forwarding of traps over NBI by its own filter set-
tings.
Additionally the SNMP Agent provides periodic notifications (for example, heartbeat or alarm summa-
ry information) as well as management information on NMS demand, for example, after receiving re-
quests for synchronization of alarms or topology data.
The NMS management capability is described in the next sections according to the following functional
groups:
In addition, the document describes the main use cases (including their related data flows) and the
layout of current MIB modules.
The provision of harmonized NBI features for different domains and network technologies strongly de-
pends on the capabilities of the managed NEs (for example, different alarm parameters might not be
present in the notifications generated by some NEs), of the underlying NetAct platforms and the con-
figurable facilities offered by the SNMP Agent itself.
Despite of its benefits about common management capabilities, the following generic approaches
must be taken into account:
• Domain-specific expectations about the granularity of the supported object model. Also in the case
of a very rough network resource modeling (for example, supporting only objects representing
NetAct and the managed NEs), significant alarm information generated in NEs is fully available in
the traps forwarded on SNMP-based NBI, that is, NMS operators are able to identify the faulty NE
resource.
• Domain or technology specific management information. The use of domain-specific
probableCause values is defined in well-established standards, for example, ITU-T, 3GPP or
TMF.
All registered NMS communicating with NetAct uses the same SNMP NBI version.
and topology synchronization, triggered by the requesting NMS, an unambiguous NMS identifier
nbiRequestingNmsName is needed.
For modifying any filter related configurations defined in the nms_conf.xml file, the NMS identifier
nbiRequestingNmsName is not needed in the SET request. The destination data in the NMS
registration table (see NMS registration in SNMP Agent) are used by the SNMP Agent in the
response.
More details about the used syntax are available in the MIB definitions.
Note:
1. The second request of another type than the first (still in execution) request can be
processed in parallel to the first one.
2. The second request of the same type as the first (still in execution) request is,
depending on the type, either rejected or queued (that is, the immediate response
sent to the NMS originating the second request means only acceptance) and
processed after the execution of the first one is finished.
• Two SNMP NBI versions may use the same MIB version, but can have different
behaviors towards the external NMS (for example, if some functional restrictions in the
older version are removed).
The NMS operator is able to retrieve the current time period used for the generation of heartbeat notifi-
cations and also to change this value. For more information, see Setting time period for heartbeat noti-
fications.
In case of multiple NMS configurations, the same heartbeat period is valid for all active northbound in-
terfaces.
A change of the heartbeat period triggered by one NMS is notified to all registered NMS. The new peri-
od value is valid for all NMS beginning in the new settings.
These procedures are transparent to the NBI, that is, the SNMP Agent forwards to all registered NMS
only the differences between the old and the new network conditions (as far as such notifications are
not filtered out according to current NBI filter settings).
For example, during the NetAct-NE alignment, NetAct compares the new and the old alarm condition
and handle it as follows:
• For each alarm present in the new alarm condition but not in the old one, NetAct forwards an ac-
tive alarm trap to the registered NMS.
• For the same unchanged alarm (that is, no change of acknowledgment state or other parameters)
present in both the new and old alarm conditions, no event report is sent towards NMS.
• For each alarm present in the old alarm condition but not in the new one, an appropriate cleared
alarm report (and, if necessary, an additional acknowledged alarm) is generated by NetAct and
forwarded to NMS.
Failures at the NetAct-NE interfaces followed by NetAct-NE synchronization do not lead to duplicated
alarm reports towards NMS.
The registration information for each NMS is pre-configured in NetAct and contains the following infor-
mation:
Note:
• If NAT is needed for the communication between NMS and NetAct, the standard
snmpTargetAddrTable contains the IP address and trap port number of the related NAT
system.
• In case one NMS (SNMP Manager) does not support the filter setting via NBI, the values of the
filter settings may be pre-configured according to default values.
• At every restart and also periodically the SNMP Agent evaluates the pre-configured and
persistently stored registration information and populates both the snmpTargetAddrTable and
nmsRegistrationTable accordingly. For example, if a further NMS communicates with NetAct,
the registration configuration is enhanced locally in NetAct. And subsequently, an additional table
row is generated without restarting the SNMP Agent. In the nmsRegistrationTable, one
NMS can only change its filter constructs related to forwarding of notifications, but not any other
parameters (see Filter settings as well).
• startUpNotification: the SNMP Agent has been (re-)started and successfully initialized.
– If the configurable parameter onlyFMsupported is set to true (default value), NMS uses
the notification as a signal to trigger only synchronization of alarms.
– If the configurable parameter onlyFMsupported is set to false, NMS uses the notification
as a signal to trigger both synchronization of alarms and topology data.
Both the startUpNotification and shutDownNotification contain the current local NetAct
time (including time zone information) as additional data.
Also the syncRequestedNotification contains the current local NetAct time (including time zone
information) as additional data.
By retrieving the nmsRegistrationTable, each NMS is able to get the NBI version which is
currently activated on its own northbound interface.
In multi-manager configurations, currently the same NBI version (active version) is used by all regis-
tered NMS.
Using a Get request about the MIB variable heartbeatPeriod, each NMS is able to retrieve the
time period currently used by SNMP Agent to generate heartbeat notifications. At NetAct start-up time,
a default value (for example, 5 minutes) pre-configured in the SNMP Agent is used.
With a Set request using the MIB variable heartbeatPeriod, one NMS (unambiguously identified
by its requestingNmsName) may change the heartbeat period value. The change is notified to all
NMS by the dedicated changedPeriodNotification. The new time period (possible values: 0 - 60
minutes) is valid for all NMS starting in the new settings.
If the changed value is set to 0, the sending of heartbeat notifications is turned off. If the new value is
invalid, the old period is still used.
Using a Get request about the MIB variable alarmSummaryPeriod, each NMS is able to retrieve the
current time period used by SNMP Agent to generate summary notifications. At NetAct start-up time, a
default value (for example, 1 hour) pre-configured in the SNMP Agent is used.
With the Set request using the MIB variable alarmSummaryPeriod, one NMS (unambiguously
identified by its requestingNmsName) may change the summary period value. The change is notified
to all NMS using the changedPeriodNotification. The new time period is valid for all NMS
starting in the new settings.
If the new value is set to 0, the sending of alarm summary notifications is turned off. If the new value is
invalid, the old period is still used.
Note: In addition or as alternative to the periodic sending, the SNMP Agent supports the
generation and forwarding of an alarm summary notification on a dedicated NMS request.
Start Up It is generated:
Table 1:
Notification type NMS request may indicate the types of notifications to be filtered out
on NBI (this concerns all notification types forwarded over NBI).
Alarm severity NMS request may indicate the values of the parameter per-
ceivedSeverity for which the related alarms shall not be forward-
ed.
Working set Each registered higher level system can set one working set, and the
notifications can be filtered in or out according to the DNs in working
set.
To create, modify and delete working set, see Managing working sets
in Working Set Manager Help.
Operators from higher level system can set, get or modify the
working set settings from the /etc/opt/oss/global/nbisnmp/
conf/mf-conf/nms-conf.xml file.
Using Get requests, one NMS operator is able to retrieve the current filter settings from the
nmsRegistrationTable (one table row is assigned to each registered NMS).
Triggering a Set request, one NMS (unambiguously identified by its requestingNmsName) may
change the filter settings for his NBI using a combination of MIB variables eventTypeFilter,
objectInstanceFilter and severityFilter.
For a user-friendly solution the filter settings are stored persistently, for example, if SNMP Agent
(re)starts, the NMS operators do not need to define again the filter settings for notification forwarding.
Note:
• Each NMS may define the filter according to the information to be received over its SNMP-based
southbound interface. As long as NMS operators do not set a filter or in the case some third-party
NMS do not support filter setting capability over NBI, the SNMP Agent will use pre-defined filter
settings by default (nevertheless also in this case one NMS is able to retrieve the currently used
filter settings).
• The SNMP Agent may reject the NMS request if the filter setting is invalid.
• The filter settings apply for both real-time event forwarding and synchronization triggered by
setting MIB variables. If the event is retransmitted, no filtering is needed, since the buffered events
already passed the current filter checks.
To identify a potential loss of notifications on NMS level as early as possible, a mechanism providing
correct synchronization of NMS management information with the real network condition is supported
at SNMP-based NBI.
1. Sequential numbering of all forwarded notifications (alarms, alarm summary notifications, topology-
related notifications, heartbeat notifications and so on.)
The SNMP Agent assigns to each notification which passed the NMS-specific filter and is
subsequently forwarded an unambiguous sequence number (MIB variable sequenceId) on NBI.
Because the filter settings may be NMS-specific, the sequence number is maintained in the SNMP
Agent separately for each registered NMS.
If SNMP Agent (re-)starts, each sequence number is initialized (value 1) and is incremented by 1
every time a new spontaneous notification is forwarded to the related NMS. When the maximum
value 2147483647 is reached, the next value to be used is 1 again.
2. Buffering of forwarded notifications for possible retransmission needs
The SNMP Agent stores a configurable amount of the most recently forwarded notifications (which
successfully passed the filter settings) in (logically) NMS-specific event buffers.
The event buffer contains the complete trap information. Once the buffer capacity is full, the old
traps are overwritten by new ones (ring buffer approach).
If DCN outage occurs (that is, no communication between NetAct and NMS), the buffering of notifi-
cations still continues to bridge short-time communication interruptions.
Note:
1. Gap between the sequenceId values of two consecutively received spontaneous notifications
(that is, notifications went lost because of short time disturbances within the communication
network).
A dedicated retransmission request (all traps) can be triggered by SNMP Manager using a
Set request about MIB variables retransmitTrapIds and requestingNmsName. The
notifications to be retransmitted are identified as a range in format <first sequenceId>-<last
sequenceId>.
No additional filtering processing is necessary, because the buffered notifications already passed
the filter tests.
After all missing traps are sent, the endOfSequence notification indicates the number of
previously forwarded traps to the requesting NMS.
2. After the re-establishment of the communication with the SNMP Agent, the SNMP Manager checks
if there is a gap between the sequenceId value of the first received trap after communication re-
establishment and the sequenceId value of last trap received before communication interruption.
Then, SNMP Manager triggers the retransmission of the lost information.
The components of the range to be set in the variable retransmitTrapIds have the format as
follows:
• If the whole amount of data to be retransmitted is still available in event buffer, the request can
be successfully executed.
• If the lost information is (partly or completely) not available in the event buffer, the request
is rejected by SNMP Agent and SNMP Manager has to trigger subsequently overall
synchronization request (for alarms and/or topology data).
If these retransmission requests are used in multiple NMS configurations, it is ensured that the
retransmitted traps and the final endOfSequence notification are forwarded only to the requesting
NMS.
As example, the following figure emphasizes (among other aspects) the SNMP components involved
in the provision of a reliable alarm reporting on SNMP-based NBI:
In this way the NBI achieves a connectionless with guaranteed delivery communication paradigm.
For more details concerning retransmission processing, see examples in NMS requests retransmis-
sion of alarm traps.
NEs are integrated to NetAct through the Southbound Mediation, and NMSs are integrated through the
SNMP NBI. All the alarms coming from the NEs are collected through the Southbound Mediation and
saved in the Alarm database in NetAct.
• SNMP NBI forwards new (spontaneous) alarms from the Alarm database to NMSs after process-
ing and filtering. The alarm is assigned an unambiguous sequenceId for each NMS and is stored
in local Event buffer as well.
• NMS sends a set request to SNMP NBI to trigger alarm synchronization.
• NMS sends a set request with sequenceId to SNMP NBI to trigger alarm retransmission, and the
requested alarms are fetched from the local Event Buffer.
The synchronization request must be achieved to the SNMP Agent first. Then the SNMP Agent trans-
fers it to the databases.
This happens when the following MIB variables are set by a registered NMS (unambiguously identified
by its requestingNmsName):
• getAllAlarms
• getActiveAlarms
• retransmitTrapIds
• acknowledgeAlarms
• unacknowledgeAlarms
• clearAlarms
• getTopologyData
After sending the last response trap, this notification is forwarded to registered NMS and includes the
following object information:
Note:
• Depending on the NMS request type, both the sequence of returned notifications and the
endOfSequence notification is sent only to requesting NMS or to all registered NMS as follows:
Some of the described NMS requests imply an automatic update of NetAct user interface as well (for
example, updated NetAct alarm list because of NMS-triggered alarm acknowledgment).
The table at the end of this section indicates which parameters are mandatory (for one or several
alarm notification types; marked with M below) or optional (that is, configurable according to the need
of the operators and included in the optionalInformation parameter) and also the content of
different alarm notification types.
sequenceId (M) The value of this parameter indicates that the se-
quence number of the event is forwarded over
one specific SNMP-based NBI. Gaps in this iden-
tifier may indicate that the loss of events is for-
warded to the current NMS.
eventTime (M) This parameter indicates the date and time of the
alarm notification generation (which may differ
from the alarmTime defined below) as DateAnd-
Time format (including UTC offset).
alarmTime (M) This parameter indicates the date and time (in-
cluding UTC offset) of alarm generation in the
faulty network resource. The parameter must be
present in new, changed and synchronization
alarm notifications.
• NEName
• siteObjAddress
• siteObjName
• siteObjDN
• controlObjSiteAddress
• controlObjMR
• controlObjName
• probabaleCauseText
• originalSeverity
• maintenanceRegionDN
• maintenanceRegion
• originalAlarmId
• alarmCount
• enrichmentTag
• relatedDNs
Note:
NSN-SNMP-NBI-
FAULTMANAGEMENT-
MIB::nbiOptionalInformati
on="NEname=Flexi NG|
maintenanceRegion=mrname
5|
originalSeverity=indetermi
nate|
probableCauseText=High
Temperature|
maintenanceRegionDN=MRC-
5/MR-5"
NSN-SNMP-NBI-
FAULTMANAGEMENT-
MIB::nbiOptionalInformati
on="NEname=|
maintenanceRegion=mrname
5|
originalSeverity=indetermi
nate|
probableCauseText=High
Temperature|
maintenanceRegionDN=MRC-
5/MR-5"
NSN-SNMP-NBI-
FAULTMANAGEMENT-
MIB::nbiOptionalInforma
tion=""
The following table indicates which parameters are present in different alarm notification types. Leg-
end:
d Alarm parameter availability depends on configuration settings, but it may have empty val-
ue.
sequenceId y y y y y y
alarmId y y y y y y
alarmType y y y y y y
objectInstance y y y y y y
eventTime y y y y y y
alarmTime y n y n n y
probableCause y y y y y y
specificProblem c d c c d c
perceivedSeverity y y y y y y
proposedRepairAction c* n c* n n c*
additionalText c c c c c c
ackState n y n n n c
ackSystemId n c n n n c
ackTime n y n n n c
ackUser n c n n n c
clearSystemId n n n n c c
clearTime n n n n y c
clearUser n n n n c c
optionalInformation c c c c c c
The detailed information received from affected resource (and stored in NetAct alarm database) is in-
cluded in the alarm notification forwarded to NMS.
A new alarm notification contains also an unambiguous, NMS-specific sequenceId value, allowing
each NMS to easily detect a possible loss of traps. In such cases, an NMS may trigger the re-
transmission of dedicated alarms (see descriptions in NMS requests retransmission of alarm traps).
The notification contains all filterable parameters (see Filter settings) and parameters present in the
original alarm and, in addition, the so-called acknowledgment history information: acknowledgment
state, acknowledgment system identifier, acknowledgment time and operator identifier.
On SNMP NBI the correlation between the original (new) alarm and the acknowledgment or
unacknowledgment alarm notification is unambiguously recognized by the same alarmId value.
Note:
• If an alarm has not been acknowledged, it is forwarded to all NMS without acknowledgment
information.
• If an alarm has been acknowledged and then unacknowledged:
1. At the unacknowledgment time it is forwarded like an alarm report with the current
unacknowledgment information.
2. If subsequently alarm synchronization is triggered by one NMS, the alarm information contains
only the unacknowledgment information. The NMS operator recognizes, that this alarm has
been acknowledged first and unacknowledged then.
This notification contains all parameters that are filterable and present in the original alarm notification,
while the perceivedSeverity value is set as cleared. In addition, the values of the parameters
clearTime, clearSystemId and clearUser indicate the time, the management system identifier
and (about manual clearing) the user identification.
At SNMP NBI the correlation between the original (new) alarm and the cleared alarm is recognized by
the same alarmId value.
Cleared alarms generated in managed NEs without corresponding active alarms in NetAct alarm data-
base are not forwarded to NMS.
Providing a quick overview about the network condition, the summary information facilitates the verifi-
cation of alarm synchronization status between each NMS and NetAct.
The alarmSummaryNotification contains the NetAct current time, the state of NetAct
synchronization with regard to the managed NEs, and the number of current alarms sorted according
to their severity values (this includes the cleared but not yet acknowledged alarms). Because the
notification takes into account the available NMS-specific filter settings defined for event forwarding,
the contents is NMS-specific.
By default, the time period for sending alarm summary notifications is pre-configured in NetAct. In
addition, NMS operators have the capability to change the time period value using the MIB variable
alarmSummaryPeriod (see Setting time period for alarm summary notifications and use case
descriptionNMS sets time period for sending alarm summary notifications).
If the variable is set to 0, the periodic sending of the summary notification is turned off.
Note:
• The provision of periodic alarm summary information is especially useful for simple third-party
SNMP Managers, which does not support more sophisticated mechanisms (like retransmission of
traps described in Retransmission of lost traps) for discovering lost traps.
• For multiple NMS, the same alarm summary time period value is valid for all active northbound
interfaces. This periodic notification does not replace the sending of heartbeat notifications, and is
usually generated in shorter time intervals.
• When the NMS is configured with working set filter, the number of current alarms in the alarm
summary notification might be affected.
The detailed changed information received from affected resource (and stored in NetAct alarm data-
base) is included in the alarm notification forwarded to NMS.
The notification contains all parameters that are filterable (see Filter settings) and present
in the original alarm, and the so-called changed information such as specificProblem,
perceivedSeverity.
On SNMP NBI the correlation between the original (new) alarm and the changed alarm notification is
unambiguously recognized by means of the same alarmId value.
The NMS-specific filter settings about real-time event reporting apply also for the alarm synchroniza-
tion response, independently of the used option.
It is possible that some of the alarms returned because of a synchronization request have been
already sent to NMS at the time of event occurrence. To avoid alarm duplication, the NMS has to
check whether the alarmId value in the synchronization trap is already available in its alarm list and.
In this case, check whether an update of its alarm information is needed.
For triggering alarm synchronization the MIB variables getActiveAlarms or getAllAlarms are
set either to the value all or to a DN identifier of a NE (in the latter case only alarms originated in the
indicated network element are returned).
The SNMP Response PDU indicates either the rejection or the acceptance of the NMS request, but
not the successfully completion of the initiated job.
Once all required alarm traps (according to alarmSyncNotification syntax) have been forwarded
to NMS, an endOfSequence notification (which includes the number of previously forwarded traps) is
sent to the requesting NMS.
If synchronization requests are used in multiple NMS configurations, the alarm notifications and the
final endOfSequence notification are forwarded only to the requesting NMS.
Generally the alarm parameters sent as a consequence of an NMS synchronization request have the
same values as in the alarm notification forwarded in real-time to the NMS. Additional parameters (for
example, when an alarm has been acknowledged in the meantime) are possible as well.
After successful processing of the NMS request, the related MIB variable is automatically reset by SN-
MP Agent to the default value (empty string).
The SNMP NBI supports the co-operative alarm acknowledgment functionality, and offers a better
alarm handling coordination between operators working on different management systems.
The feature makes distinction only between two so-called acknowledgment state values, that is,
unacknowledged and acknowledged. An acknowledged alarm is still managed on NBI only if it is
still active. Generally NetAct alarm list contains both active (not yet acknowledged, acknowledged or
unacknowledged) and cleared (not yet acknowledged or unacknowledged) alarms.
When alarms are acknowledged on NMS level (that is, manually by NMS operator or automatically
by NMS system) the MIB variable acknowledgeAlarms is set to true (1) and the Set request
contains also the list of alarm identifiers (alarmIdList), the operator identifier (ackUser, see
use case in NMS triggers alarm acknowledgment), the requestingNmsName and (optionally) the
ackSystemId.
Note: In the Alarm List of Monitor application, the value in the column Acknowledgment
User is combined with the ackUser and ackSystemId, which is separated by |, and the
maximum length is 64.
The alarm database in NetAct and also the presentation of acknowledged alarms on NetAct user
interface are subsequently updated. After successful processing the NMS request, the SNMP
Agent generates for each alarm a corresponding ack changed notification (see Notification about
alarm acknowledgment or unacknowledgment) and forwards them (under consideration of current
filter settings) to the registered NMS (including the requesting NMS). Finally an endOfSequence
notification (which includes the number of previously forwarded notifications) is sent to the registered
NMS.
If one or several alarms are acknowledged by mistake, one NMS operator could want to un-
acknowledge them, that is, the alarm(s) are set in the state unacknowledged using the MIB variable
unacknowledgeAlarms in a similar way.
If the NMS request contains also identifiers of alarms not available anymore in the NetAct alarm
database, the NMS receives less notifications than the number of alarms indicated in the request. In
such cases the parameter processingStatus in the endOfSequence notification has the value
partialSuccess.
After successful processing of the NMS request, the related MIB variable is automatically reset by
SNMP Agent to false(2).
Note: In the Alarm List of Monitor application, the value in the column Cancel User is
combined with the clearUser and clearSystemId, which is separated by |, and the
maximum length is 64.
After successful processing the NMS request, that is, automatic update of alarm list and of the related
NetAct user interface, the SNMP Agent generates for each cleared alarm a corresponding clear
alarm notification (see Notification about alarm clearing) and forwards them (under consideration of
current filter settings) to all registered NMS (including the requesting NMS). Also an endOfSequence
notification (which includes the number of previously forwarded traps) is sent to the registered NMS.
If a manually cleared alarm was previously already acknowledged, it is automatically removed from the
NetAct alarm list. If an NE alarm is cleared manually by mistake (that is, the alarm is still active in the
network), the alarm will be received again after the next alarm synchronization between NetAct and
the related NE.
It is the responsibility of the NMS operator to clear manually only alarms, which are not active anymore
in the network but which have not been cleared automatically, for example, because of the loss of the
related cleared alarms.
When the NMS request contains also identifiers of alarms not available anymore in the NetAct alarm
database, the NMS receives less notifications than the number of alarms indicated in the request. In
such cases the parameter processingStatus in the endOfSequence notification has the value
partialSuccess.
After successful processing of the NMS request, the related MIB variable is automatically reset by
SNMP Agent to false(2).
• An NMS operator needs urgent check of the alarm situation and does not want to wait until the
periodic summary notification is received.
• The periodic sending of alarm summary notification has been previously turned off.
In such cases the MIB variable getAlarmSummary is set to value true (1).
After successful processing the NMS request (that is, forwarding of alarmSummaryNotification
to the requesting NMS), the MIB variable is automatically reset by SNMP Agent to the default value
false (2).
Note:
• The current filter settings towards external NMS apply for these alarm notifications as well.
• The sequenceId values are temporarily stored in event buffer and can be retransmitted when
they have been lost (according to Reliable event reportingabove).
• The short description related to the problem of the current Agent is conveyed in the parameter
additionalText of the alarm notification and a more detailed description is present in the
Troubleshooting documentation
• NMS is able to retrieve topology-related information about the managed network resources
• Configuration changes are sent under consideration of NMS-specific filter settings as real-time
event reports (object creation, object deletion) to all registered NMS.
eventTime (M) This parameter indicates the date and time of the
create notification generation as DateAndTime
format (including UTC offset).
• NeName
• siteObjAddress
• siteObjName
• siteObjectDN
• maintenanceRegion
• vendor
• version
• alarmCount
• enrichment tag
eventTime (M) This parameter indicates the date and time of the
delete notification generation as DateAndTime
format (including UTC offset).
On reception of this deletion notification, NMS deletes the related instance and also all related alarms
from its database.
• To the value all: the network topology information about all alarming objects is returned
(mandatory)
• To the DN of a baseObject value: the topology information of the baseObject itself and all
contained managed resources is returned (optional).
Generally the feature can be triggered by NMS operators at any time, but it is strongly recommended
at least in the following cases:
The response is a sequence of topologySync notifications, which contain the same parameters as
the object creation notification.
Once all the notifications have been forwarded, the SNMP Agent sends to the requesting NMS an
endOfSequence notification, which includes the number of previously forwarded traps.
If a synchronization request is used in multiple NMS configurations, it is ensured that the topology-
related traps and the final endOfSequence notification are forwarded only to the requesting NMS.
After successful processing of the NMS request, the MIB variable is automatically reset by SNMP
Agent to the default value (empty string).
4 Feature summary
The following tables contain a summary of SNMP NBI features from a functional point of view:
Set filter for notifica- Set MIB variables: Filter settings are NMS-specific and for this
tions eventTypeFilter, purpose the nmsRegistrationTable is used,
while each row in the table contains the current
objectInstanceFilter,
severityFilter settings about the eventTypeFilter, the
and additionally objectInstanceFilter and (only for alarms)
the NMS identifier the severityFilter.
(requestingNmsName)
The filter settings apply for spontaneous notifica-
tions, because of notifications sent as response to
previous NMS requests and in synchronization re-
sponses.
Retransmit traps Set MIB variable re- Requested traps still available in event buffer are
from event buffer transmitTrapIds and forwarded as a sequence of single traps to the re-
additionally the NMS questing NMS. No additional filter check needed,
unambiguous identifier because all traps in event buffer already passed
(requestingNmsName) this check.
Alarm synchroniza- Set MIB variable getActiveAlarms: Only active alarms (that
tion using FM plat- getActiveAlarms or is, alarms whose perceived severity value is not
form data getAllAlarms and clear) generated by all resources or by specified
additionally the NMS managed elements (including NetAct) are returned
unambiguous identifier as a sequence of single traps.
(requestingNmsName)
getAllAlarms: alarms generated by all
resources or by specified managed elements are
returned as a sequence of single traps.
NMS triggers alarm Set MIB variable ac- The variable list in the Set request contains in ad-
acknowledgment/un- knowledgeAlarms dition the list of alarmId values for the alarms to
acknowledgment or unacknowl- be acknowledged or unacknowledged (using the
NMS triggers alarm Set MIB variable The variable list in the Set request contains in addi-
clearing clearAlarms to val- tion the list of alarmId values for the alarms to be
ue true(1), the user cleared (using the alarmIdList object) and the
identifications and also NMS user identification (clearUser).
the NMS unambiguous
identifier (request-
ingNmsName)
NMS triggers im- Set MIB variable getA- The variable value is set to value true (1). After
mediate sending of larmSummary and ad- the processing of the request is finished, the vari-
alarm summary noti- ditionally the NMS un- able is set by SNMP Agent to value false (2).
fication ambiguous identifier
(requestingNmsName)
NE topology data Set MIB variable get- Topology data of managed resources (either about
synchronization us- TopologyData and all alarming objects or only a subtree defined by
ing CM platform additionally the NMS means of a baseObject) are returned as a se-
unambiguous identifier quence of single traps.
(requestingNmsName)
Set period for send- Set MIB variable The heartbeat period is valid for all registered
ing heartbeat notifi- heartbeatPeriod and NMS.
cations additionally the NMS
unambiguous identifier
(requestingNmsName)
Set period for send- Set MIB variable alar- The summary period is valid for all registered NMS.
ing summary notifica- mSummaryPeriod and
tions additionally the NMS
unambiguous identifier
(requestingNmsName)
Parameters clearSystemId,
clearUser, clearTime are not present.
Parameters clearSystemId,
clearUser, clearTime have well-
defined (or empty) values.
shutDownNotification
syncRequestedNotification
Note:
• In each exchanged message only the relevant parameters for the described use case
are listed. Behind every shown trap the specific trap type is indicated.
• For readability, the prefix nbi used in the MIB definitions are intentionally omitted in the
diagrams.
1. NMS sends set request to SNMP NBI to set objectInstanceFilter to NSNNetwork-1/RNC-2 and
severityFilter to 48.
2. SNMP NBI sends response to NMS, and from now on, SNMP NBI will filter out all alarms from
RNC-2 whose severity is warning or minor for this NMS.
The current filter applies for all related NBI mechanisms: real-time event forwarding, event retransmis-
sion (implicitly, since all notifications in event buffer already passed the filter check) and synchroniza-
tion requests.
5.2 NMS sets time period for sending alarm summary notifications
In this example, NMS1 changes the time intervals for sending the alarm summary notifications from 10
hour to 15 hours.
1. NMS sends get request to SNMP NBI to retrieve the alarm summary period.
2. SNMP NBI sends response to NMS with the alarm summary period value 10.
3. NMS sends set request to SNMP NBI to set alarm summary period value to 15.
4. SNMP NBI sends changedPeriodNotifications to all registered NMS to inform that the new
alarm summary period is changed to 15.
A similar approach is used in case one: NMS operator changes the time period for sending heartbeat
notifications.
• It is possible that some of the previously forwarded (spontaneous) alarms have been
acknowledged and cleared in the meantime (that is, they are not managed in the NetAct’s alarm
database), thus gaps between the alarmId values of consecutive alarms may occur.
• During the alarm synchronization the forwarding of new occurred alarms towards the requesting
NMS is supported. The NMS application is able to distinguish between alarms sent as response to
the previous request and spontaneous alarm traps because of the different trap OIDs.
• At the end of synchronization, a dedicated endOfSequence trap indicates the number of sent
alarms related to the synchronization request. This is a means for the requesting NMS to check
the correct reception of all forwarded traps.
1. NMS sends Set request with variable getActiveAlarms = all to SNMP NBI to trigger syn-
chronization for all active alarms.
2. SNMP NBI sends response to NMS.
Recognition of missing traps and lost information are still available in event buffer.
In this example (for example, short time NMS-NetAct communication interruption or network
problems), an NMS interested only in alarm traps recognizes a gap between the sequenceId values
of two consecutive received alarms and the needed information is still available in event buffer.
The NMS uses a Set request to indicate the sequenceId values of the alarms to be retransmitted.
The related alarms are set as acknowledged in NetAct alarm database. Subsequently corresponding
acknowledgment traps are generated and forwarded to both NMS1 and NMS2 (under consideration of
current NMS-specific filter settings).
The sequence diagram below emphasizes that for each alarm the acknowledgment traps are sent to
all registered NMS with the same alarmId value but NMS-specific sequenceId values.
Note: For example, the new alarm occurrence with alarmId = 1238 was filtered
out, also the related alarm acknowledgment trap is not forwarded on NBI, that is, the
endOfSequence notification for NMS2 would indicate that only two traps have been sent.
Note:
• The list below may be extended during the implementation phase of the current version
and also for future NBI versions. In any cases, the extensions should be done in a
backwards compatible way.
• All configurable parameters are valid SNMP-Agent wide, as far as NMS-specific settings
are not explicitly mentioned.
• Community string and associated access rights. Further security-related parameters need to be
configurable in case of SNMPv3 support.
• SNMP version (SNMPv2c or SNMPv3) used at northbound interface.
• IP addresses and trap destination port numbers used by SNMP Agent for bi-directional
communication with all registered NMS.
• Symbolic names of used NMS (needed for registration).
• Default time period the SNMP Agent checks the NMS registration information (pre-configured in
NetAct).
• Default NMS-specific filter settings about notification forwarding (needed when NMS operators do
not set filter or when some third-party SNMP Managers do not support filter settings capability at
NBI).
• NMS-specific filter settings locally defined in NetAct.
• Default time period for sending heartbeat notifications.
• Default time period for generation of periodic alarm summary notifications.
• Enabling or disabling forwarding different traps by setting filters.
• Customer configurable content of the parameter optionalInformation in alarm and topology
notifications.
• Separator character between the sub-string conveyed in the parameters additionalText or
optionalInformation.
8 MIB structure
The mediation application function provided by the SNMP Agent converts the system dependent view
of the lower level NetAct-NE interfaces into the system-independent view at NBI.
The object model exposed by SNMP-based northbound interface is formally described in the SNMP
NBI Management Information Base (MIB). The Nokia-specific MIB for the SNMP NBI is defined ac-
cording to the following naming tree:
• Registration MIB, which contains the unambiguous top-level OID (nbiNetActSnmp) for all SNMP
NBI definitions.
• Textual conventions MIB, which contains standard conformant textual conventions for the objects
used in the following modules.
• Common functions MIB, which contains object definitions for NBI functions not related to a dedi-
cated management area (fragment).
• Fault management MIB, which contains object definitions for NBI functions about the management
of alarms.
• Topology management MIB, which contains object definitions for NBI functions about the manage-
ment of topology data.
This modular approach allows for a seamless extension of the MIB in case further NBI management
areas will be provided in future versions.
The MIB definitions take into account (among others) the RFC 4181 Guidelines for Authors and Re-
viewers of MIB Documents about the OID layout:
• The value of the module-identity invocation is used as a sub-tree under which other object
identifier values assigned within the module are defined.
• Immediately below the module-identity value there are three separate branches dedicated
(respectively) to notification definitions, object definitions and conformance definitions.
Furthermore the conformance branch is subdivided into separate sub-branches for compliance
statements and object or notification groups. This structure is illustrated in the picture below,
where the numbers in parentheses are the sub-identifiers assigned to the branches.
9 List of abbreviations
CM Configuration Management
FM Fault Management
IP Internet Protocol
NE Network Element
NM Network Management