Applications Manager provides out-of-the-box performance metrics and helps you visualize the health and availability of an Oracle NoSQL server farm. Database administrators can login to the web client and visualize the status and Oracle NoSQL performance metrics.
Supported Versions: Oracle Nosql 12cR1.3.5.2 and below.
Using the REST API to add a new Oracle NoSQL monitor: Click here
Follow the given steps to create a Oracle NoSQL server monitor:
Oracle NoSQL Database allows Java Management Extensions (JMX) or Simple Network Management Protocol (SNMP) agents to be optionally available for monitoring, in addition to the native monitoring provided by the Admin CLI and the Admin Console. These agents provide interfaces on each storage node that allow management clients to poll them for information about the status, performance metrics, and operational parameters of the storage node and its managed services, including replication nodes and admin instances. Both these management agents can also be configured to push notifications about changes in the status of any of the services, and for violations of preset performance limits. The JMX interface can be enabled in either the Community Edition or the Enterprise Edition. To use SNMP, however, you must have the Enterprise Edition. The JMX service exposes MBeans for the three types of components. These MBeans are the java interfaces StorageNodeMBean, RepNodeMBean, and AdminMBean in the package oracle.kv.impl.mgmt.jmx. For more information about the status reported for each component, see the javadoc for these interfaces.
The same information that is reported via JMX can also be reported through SNMP. In this case, the information is organized according to the Management Information Base (MIB) named OracleNosqlMIB, which is included with the Enterprise Edition, in the file lib/nosql.mib.
Both the SNMP and JMX agents in NoSQL Database are read-only interfaces and allow you to poll the storage nodes for information about the storage node and about any replication nodes or admins that are hosted on the storage node. The available information includes service status (such as, RUNNING, STOPPED etc.), operational parameters, and performance metrics. SNMP and JMX traps/notifications are also delivered for particular events. Notifications are sent for every service status state change; and for violations of performance limits.
Monitoring can be enabled on a per-storage node basis in two different ways:
You can specify that you want to enable JMX or SNMP in the storage node's boot configuration file. Usually, these files are created by using the makebootconfig utility, which has the following options to control these features:
[-mgmt {snmp|jmx|none} -pollport <snmp poll port>]
-traphost <snmp trap/notification hostname>]
-trapport <snmp trap/notification port>]
You can still enable JMX or SNMP after a store is deployed, by changing the storage node parameters "mgmtClass", "mgmtPollPort", "mgmtTrapHost", and "mgmtTrapPort". Similar to configuring via makebootconfig, the "mgmtPollPort", "mgmtClass", "mgmtTrapHost", and "mgmtTrapPort" are used only for SNMP; and the parameter "mgmtPollPort" must be set when enabling SNMP.
The value of the "mgmtClass" parameter may be one of the following class names:
oracle.kv.impl.mgmt.jmx.JmxAgent
oracle.kv.impl.mgmt.snmp.SnmpAgent
oracle.kv.impl.mgmt.NoOpAgent
For example, you could issue the following command in the Admin CLI to enable SNMP on a storage node:
plan change-parameters -service sn1 -wait -params \ mgmtClass=oracle.kv.impl.mgmt.snmp.SnmpAgent \ mgmtPollPort=5002 mgmtTrapHost=192.168.26.42 mgmtTrapPort=32767
To view the NoSQL Database JMX Mbeans in a monitoring tool such as JConsole, connect using the hostname and registry port for each Storage Node that you would like to view. This is the port that is named by the "-port" argument to the java -jar kvstore.jar makebootconfig command. It is also used in the deploy-sn plan during configuration. The documentation examples use 5000 as the registry port.
For example, you would specify localhost:5000 to the JConsole Remote Process connection box in the New Connection tab.
Oracle NoSQL Monitoring Metrics:
Attribute Name | Description |
---|---|
Admin ID | Number of an Admin instance. |
HttpPort | The port number used by the Admin Console web application. |
LogFileCount | Number of log files that are kept. |
LogFileLimit | The maximum size of log files. |
Master |
Whether this Admin is the master among Admin instances.
|
PollPeriodMillis | The polling period for collecting metrics. |
ServiceStatus | The reported service status of the Admin. |
Attribute Name | Description |
---|---|
Capacity | The number of Replication Nodes that can be hosted on this Storage Node. |
AdminHttpPort | The http port used by the Admin Console web application. |
HostName | The name of the network interface used for communication between Replication Nodes |
HAPortRange | The range of port numbers available for assigning to Replication Nodes that are hosted on this Storage Node. |
HostingAdmin | True if this Storage Node hosts an Admin instance |
MemoryMB | The amount of memory known to be available on this Storage Node, in megabytes. |
MountPoints | A list of file system mount points on which Replication Nodes can be deployed |
NumCPUs | The number of CPUs known to be available on this Storage Node. |
RegistryPort | The port number of the Storage Node's RMI registry. |
RootDirPath | The pathname of the store's root directory |
SnId | The StorageNodeId number of this Storage Node. |
StoreName | The configured name of the store to which this Storage Node belongs. |
Attribute Name | Description |
---|---|
CacheSize | The size of the BDB-JE cache, in MBytes. |
CollectEnvStats | If true, then the underlying BDB-JE subsystem will dump statistics into a local .stat file. |
CommitLag | Average commit lag(in milliseconds) for a given replication node's update operations during a given time interval. |
RepNodeId | The RepNodeId, in its String form |
CommitLagThreshold | Average commit lag (in milliseconds) for a given replication node during a given time interval exceeds the value returned by this method, a notification event will be sent to any parties that have registered interest. |
CumulativeEnd | The timestamp at the end of a cumulative measurement period. |
CumulativeLatAvg | The average latency for single operations since service startup. |
CumulativeLatMax | The highest latency measured for single operations since service startup. |
CumulativeStart | The timestamp at the beginning of a cumulative measurement period |
CumulativeThroughput | Single operations per second since service startup. |
CumulativeTotalOps | The total number of single operations since service startup. |
HeapMB | The size of the Java heap for this Replication Node, in MB. |
IntervalLatAvg | Average Latency of Operations in the given Interval |