Event- and alert type hierarchy¶
NAV events and alerts are organized into a type hierarchy. While NAV’s backend monitoring processes usually generate events, the Event Engine is responsible for deciding what to do with those events. In most cases, events are translated into a corresponding alert by the Event Engine.
Most alerts are stateful, i.e. they can be viewed as an incident that has a start time and an end time. If the end time is set to an infinite value, the alert is considered unresolved (also known as open).
Some alert types indicate the beginning of a new alert state, while some alert
types indicate the closure of an existing alert state. E.g. for boxState
alerts, a boxDown
alert indicates that a new stateful incident has occured:
A device is no longer responsive. On the other hand, a boxUp
alert will
cause an existing boxState
-type alert to be resolved.
All legal event- and alert-types are registered in the NAV database (and can thus be extended, even by the user, if need be). The following are the event- and alert types that come pre-defined with NAV:
boxState events¶
Tells us whether a network-unit is down or up.
Alert type name |
Description |
---|---|
|
Warning sent before declaring the box down. |
|
Warning sent before declaring the box in shadow. |
|
Box declared down. |
|
Box declared up. |
|
Box declared down, but is in shadow. |
|
Box declared up from a previous shadow state. |
serviceState events¶
Tells us whether a service on a server is up or down.
Alert type name |
Description |
---|---|
|
http service not responding. |
|
http service responding. |
moduleState events¶
Tells us whether a module in a device is working or not.
Alert type name |
Description |
---|---|
|
Warning sent before declaring the module down. |
|
Module declared down. |
|
Module declared up. |
thresholdState events¶
Tells us whether the load has passed a certain threshold.
Alert type name |
Description |
---|---|
|
Threshold exceeded. |
|
Value below threshold. |
linkState events¶
Tells us whether a link is up or down.
boxRestart events¶
Tells us that a network-unit has done a restart
Alert type name |
Description |
---|---|
|
The IP device has coldstarted |
|
The IP device has warmstarted |
info events¶
Basic information
Alert type name |
Description |
---|---|
|
Mismatch between sysname and dnsname. |
|
Serial number for the device has changed. |
|
Mac appeared on port |
notification events¶
Notification event, typically between NAV systems
deviceActive events¶
Lifetime event for a device
deviceState events¶
Registers the state of a device
Alert type name |
Description |
---|---|
|
The device is now in operation with an active IP address. |
|
The device is now in operation as a chassis module. |
|
RMA event for device. |
|
The device has been found as a module. |
|
The device has been found as a chassis. |
|
The device has been found as a power supply. |
|
The device has been found as a fan. |
deviceNotice events¶
Registers a notice on a device
Alert type name |
Description |
---|---|
|
Error situation on device. |
|
Software upgrade on device. |
|
Hardware upgrade on device. |
|
Firmware upgrade on device. |
maintenanceState events¶
Tells us if something is set on maintenance
Alert type name |
Description |
---|---|
|
Box put on maintenance. |
|
Box taken off maintenance. |
apState events¶
Tells us whether an access point has disassociated or associated from the controller
Alert type name |
Description |
---|---|
|
AP associated with controller |
|
AP disassociated from controller |
snmpAgentState events¶
Tells us whether the SNMP agent on a device is down or up.
Alert type name |
Description |
---|---|
|
SNMP agent is down or unreachable due to misconfiguration. |
|
SNMP agent is up. |
chassisState events¶
The state of this chassis has changed
Alert type name |
Description |
---|---|
|
This chassis is no longer visible in the stack |
|
This chassis is visible in the stack again |
aggregateLinkState events¶
The state of this aggregated link changed
Alert type name |
Description |
---|---|
|
This aggregate link has been degraded |
|
This aggregate link has been restored |
psuState events¶
Reports state changes in power supply units
Alert type name |
Description |
---|---|
|
A PSU has entered a non-OK state |
|
A PSU has returned to an OK state |
fanState events¶
Reports state changes in fan units
Alert type name |
Description |
---|---|
|
A fan unit has entered a non-OK state |
|
A fan unit has returned to an OK state |
bgpState events¶
The state of this BGP peering session changed
Alert type name |
Description |
---|---|
|
This BGP peering session is down |
|
This BGP peering session has been established |
|
This BGP peering session is administratively down |