What is Active Discovery?
Active Discovery is the process by which LogicMonitor determines all of the similar components of a particular type on given system what we call datasource instances. The output of an Active Discovery process is one or more instances upon which we can collect a particular type of data.
Active Discovery is extremely valuable for ensuring that monitoring keeps up to date with changes to your environment. For instance, Active Discovery ensures that each new VM created on your virtualization stack is monitored, or that each new volume added your storage system is monitored for utilization and latency. Without Active Discovery, you'd have to remember to update your monitoring instances every time you made a change to your production environment. Ain't no one got time for that.
How frequently does Active Discovery look for instances?
Datasources with Active Discovery enabled have a discovery schedule that defines the time between Active Discovery runs. The discovery schedule varies per datasource; typically items that change less frequently (such as the number of fans or CPUs in a system) have a discovery schedule of once a day, whilst objects that are likely to change more frequently (such as the volumes on a NetApp disk array) typically have a discovery schedule of several times per hour.
Active Discovery will run at the following times:
- if N minutes have elapsed since the last run, where N is the Active Discovery Schedule.
- if any instance stops reporting data, Active Discovery is triggered immediately. This lets LogicMonitor determine if the instance is no longer reporting data because it is no longer present, or if it is failing, and needs to be alerted on.
- if any of the properties of a device (such as snmp.community) are changed.
- If initiated manually via the Run Active Discovery action. It is sometimes useful to trigger Active Discovery if you have added a new object on a device, and want to be sure LogicMonitor picks it up immediately, without having to wait for the next Active Discovery cycle.
How does Active Discovery find instances?
LogicMonitor's Active Discovery system supports a variety of mechanisms for querying a device or system about the objects it has available for monitoring. These include:
- CIM – query a Common Information Model class & property
- ESX – query the VMware ESX API
- HTTP – query an HTTP/HTTPS URL
- JDBC – actuate a database query
- JMX – query a specific Java Management Extensions path
- MONGO – use the MongoDB API
- NETAPP – use the NetApp API
- PERFMON – query a Windows Perfmon class
- PORT – connect to specific TCP port(s)
- SCRIPT – use the output of a script
- SNMP – actuate an SNMP query
- WMI – query a Windows WMI class
- XEN – use the Citrix XenServer API
- CLOUDWATCH – query AWS CloudWatch metrics
What information does Active Discovery gather?
Active Discovery retrieves the following information for each instance it finds:
Instance Name: The descriptive name of the instance (e.g. Fastethernet0) also known as the "ALIAS" as displayed in the device tree.This value must be unique for the device/datasource combination.
Instance ID: The unique id of the instance, used as the variable to identify this instance when querying the device for data. For example, the variable part of an SNMP OID tree, or the volume ID in a storage system. This is also known as the "WILDVALUE".
Instance Description: An optional description of the instance, which is displayed along with the instance name in the device tree.
Instance Properties: An optional set of key-value pairs that gather static data about this instance. These are analogous to device properties, but collected on a per-instance basis. Once collected, instance properties are displayed in the UI, and can be used as a key to group instances together in complex datapoint calculations. The following properties are commonly collected as instance properties:
- Serial numbers for each FRU discovered in a switch chassis, or drives in a storage system
- Metadata for each VM hosted by a Hypervisor (CPU Count, Memory Allocation, virtual NIC count, guest OS, etc.)
- Port speed for each network interface
Note that Instance Properties are available only with SCRIPT, SNMP, and WMI Active Discovery mechanisms, and on Collector versions 22.060 and higher.