Apache Cassandra Monitoring
Overview
Applications Manager provides out-of-the-box performance metrics and helps you visualize the health and availability of your Apache Cassandra NoSQL database.
Creating a new Cassandra DB monitor
Supported Versions: Apache Cassandra 2.0 and above.
Prerequisite:Click here to know about the necessary prerequisites to monitor Apache Cassandra database.
Using the REST API to add a new Cassandra monitor:Click here
To create a Cassandra database server monitor, follow the steps given below:
- Click on New Monitor link.
- Select Cassandra.
- Specify the Display Name of the Cassandra monitor
- Enter the HostName or IP Address of the host where Cassandra server runs
- Enter the JMX Port where the server is running. By default, it will be 7199 or else check in cassandra-env.sh file for the JMX_PORT.
- If you want to discover only this node, disable the option Discover all nodes in the Cluster. By default, it is enabled which means all the nodes in the cluster are discovered by default
- Check the Is Authentication Required field, to give the JMX credentials to be used to connect to the Cassandra server. Now enter the UserName and Password of the credential
- Specify the Polling Interval in minutes
- If you are adding a new monitor from the Central Server, select a Probe Server.
- Choose the Monitor Group with which you want to associate the Cassandra server to, from the combo box (optional). You can choose multiple groups to associate your monitor
- Click Add Monitor(s). This discovers the Cassandra server from the network and starts monitoring it
Note:
In case you are unable to add the monitor even after enabling JMX, try providing the below argument:
-Djava.rmi.server.hostname=[YOUR_IP]
Monitored Parameters
Click on the individual monitors listed to view the following information in the below tabs:
Monitor Information
This tab displays the basic information like the name of the server, host name and the JMX port where the server is running. Additional parameters like Token, Gossip Active, Load, Generation No, Uptime, DataCenter, Rack Name are also displayed. These are the equivalent of the Nodetool result got using the command 'nodetool.bat/sh -h -p info'
Performance
| Parameters | Description |
|---|
| Availability and Performance History | Displays the availability and performance history for the last six hours. |
| CPU Usage (available till version 2.1.13) | Specifies the usage of CPU in percentage. |
| Memory Utilization | Specifies the used, free and total memory of the server in mega bytes |
| Exceptions (available till version 2.1.13) | Count of various exceptions such as not found, invalid, unavailable, timed out, etc. |
| Storage Stats | Specifies the storage load in GB. |
| Operation Stats (available till version 2.1.13) | Number of range, read and write operations per second since executor start. |
| Operation Latency (available till version 2.1.13) | The latency of range, read and write operations since the last time the attribute was read. |
| Bloom Filter Statistics(available from Version 2.0.9 onwards) |
| Bloom Filter False Positives | Number of false positives (Occurs when the bloom filter says a non-existential row exists) |
| Bloom Filter False Positive Ratio | Fraction of all bloom filter checks resulting in a false positive |
| Bloom Filter Disk Space Used | The size of the bloom filter files on disk |
| Memtable Statistics(available from Version 2.0.9 onwards) |
| Memtable Columns Count | Total number of columns present in memtable |
| Memtable Data Size | Amount of data stored in memtable, including column-related overhead |
| Memtable Switch Count | Number of times flushing results in the memtable being switched out |
| Client Details |
| Connected Native Clients | Number of Native Clients currently connected to the server instance |
| Connected Thrift Clients | Number of Thrift Clients currently connected to the server instance |
| Average Operational Latency (available till version 2.1.13) |
| Average Range Latency | The average latency per range operation since the last poll |
| Average Read Latency | The average latency per read operation since the last poll |
| Average Write Latency | The average latency per write operation since the last poll |
| Operational Timeouts(available till version 2.1.13) |
| Recent Timeouts | Number of read, write and range slice operation timeouts since the last poll |
Pending Tasks
| Parameters | Description |
|---|
| COMMITLOG/COMPACTION - PENDING TASKS(available till version 2.1.13) |
| Commitlog - Pending Tasks | Number of Commitlog tasks waiting in the queue to be executed. |
| Compaction - Pending Tasks | Number of Compaction tasks waiting in the queue to be executed. |
| COMMANDS/RESPONSES - PENDING TASKS(available till version 2.1.13) |
| Commands - Pending Tasks | Number of Command tasks waiting in the queue to be executed. |
| Responses - Pending Tasks | Number of Response tasks waiting in the queue to be executed. |
| Task Statistics of Thread Pools (available till version 2.1.13) | Status of various stages of thread pools like active and pending tasks, completed and blocked tasks are being displayed. These are equivalent of the Nodetool result got using the command 'nodetool.bat/sh -h <hostname> -p <portnumber> tpstats'. |
Cluster Stats
| Parameters | Description |
|---|
| Cluster Details | Specifies details on live, leaving, moving, joining and unreachable nodes. |
| Cluster Node Details | Health status and storage load (in GB) of the nodes in the cluster. |
Messages
This tab contains the Dropped Message and Message Timeout statistics:
| Parameter | Description |
|---|
| Recent Total Timeouts | Number of timeouts since last poll |
Configurations
This tab contains the configuration values of files that are part of cassandra.yaml config file.
