Help Docs

Amazon MSK monitoring

Amazon Managed Streaming for Apache Kafka (MSK) enables you to build and run applications that use Apache Kafka to process streaming data. Amazon MSK automates tasks such as hardware provisioning, software installation, cluster configuration, scaling, and monitoring, making it easier for developers and administrators to focus on building and managing their data streaming applications.

Overview

Integrating Amazon MSK with Site24x7 offers several benefits for managing and monitoring your Kafka infrastructure. Site24x7 provides real-time monitoring capabilities, allowing you to track the performance and health of your Amazon MSK Cluster, Serverless, Replicator, Connect, Broker, and Topic data continuously.

The integration also provides you with the following six monitors so that you can effectively monitor your Kafka environment:

  • MSK Cluster: Monitors the broker instances and topics within your cluster.
  • MSK Serverless: Gain real-time visibility into your MSK Topics.
  • MSK Replicator: Ensures timely data replication between Kafka clusters by tracking metrics like replication latency, message lag, throughput, and failure count, enabling proactive optimization for seamless data synchronization and reliability.
  • MSK Connect: Monitors the Kafka Connector performance and provides insights into critical metrics for proactive issue resolution and seamless data integration.
  • MSK Broker: Monitors your broker instances and provides insights about metrics such as resource utilization, network connectivity, and disk usage.
  • MSK Topic: Gain detailed insights into data ingestion rates, message throughput, and other metrics at each topic level.

Use case

Imagine managing an MSK cluster with numerous brokers and topics. Monitoring the health and performance of individual MSK brokers and MSK topics is crucial for minimizing the risk of service interruptions or downtime. However, this becomes particularly challenging when dealing with multiple brokers distributed across the cluster while monitoring each MSK broker and topic individually.

By integrating Amazon MSK with Site24x7, you can enhance your monitoring capabilities and gain valuable insights into crucial metrics and data at both the broker and topic levels. This integration facilitates a comprehensive understanding of the health and performance of each MSK broker instance, along with providing insights into the data ingestion rates and throughput of MSK topics. You can also identify which MSK broker instances are underutilized or overutilized, enabling you to take appropriate action to maintain high availability and reliability of your MSK clusters.

Benefits of the Site24x7-Amazon MSK integration

On integrating your Amazon MSK with Site24x7

  • Obtain comprehensive monitoring of your Kafka environments and optimize your infrastructure performance and reliability.
  • Gain insights about your monitor clusters, topics, brokers, connectors, replicators, and serverless components.
  • Ensure replicator synchronization for consistent data replication across clusters.
  • Achieve critical visibility into the data flow of all your topics and identify performance trends.
  • Monitor every Kafka broker in your cluster individually, and proactively get notified in case of abnormal behavior.

Setup and configuration

Permissions

  • kafka:ListClustersV2
  • kafka:DescribeClusterV2
  • kafka:ListNodes
  • kafka:ListReplicators
  • kafka:DescribeReplicator
  • kafkaconnect:ListConnector
  • kafkaconnect:DescribeConnector
  • kafkaconnect:DescribeCustomPlugin
  • kafkaconnect:DescribeWorkerConfiguration

Polling frequency

Site24x7 queries AWS service-level APIs per the set polling frequency (from once a minute to once a day) to collect metrics from the Amazon MSK monitor.

Supported metrics

The supported metrics for each MSK monitors are given below.

MSK Cluster

The supported metrics for MSK Cluster monitor are provided below:

Metric name Description Statistics Unit
Kafka DataLogs Disk Used The disk space used for datalogs in percentage. Maximum Percentage
Active Controller Count The number of controllers active at a given time. Minimum Count
Offline Partitions Count The number of offline partitions. Maximum Count
Client Connection Count The number of clients which are actively connected. Maximum Count
Zookeeper Request Latency Ms Mean The mean latency for Apache ZooKeeper requests. Average Milliseconds
Global Partition Count The total number of partitions in a cluster excluding the replicas. Maximum Count
Global Topic Count The total number of topics in a cluster. Maximum Count
Number of Brokers in Up state The number of Brokers in up status. Sum Count
Number of Brokers in Trouble state The number of Brokers in trouble status. Sum Count
Number of Brokers in Critical state The number of Brokers in critical status. Sum Count
Number of Brokers in Down state The number of Brokers in down status. Sum Count
Number of Topics in Up state The number of Topics in up status. Sum Count
Number of Topics in Down state The number of Topics in down status. Sum Count
Number of Topics in Trouble state The number of Topics in trouble status. Sum Count
Number of Topics in Critical state The number of Topics in critical status. Sum Count

MSK Serverless

The supported metrics for MSK Serverless monitor are provided below:

Metric name Description Statistics Unit
Number of Topics in Up state The total number of Topics monitors in Up status. Sum Count
Number of Topics in Down state The total number of Topics monitors in Down status. Sum Count
Number of Topics in Trouble state The total number of Topics monitors in Trouble status. Sum Count
Number of Topics in Critical state The total number of Topics monitors in Critical status. Sum Count

MSK Replicator

The supported metrics for MSK Replicator monitor are provided below:

Metric name Description Statistics Unit
Replication Latency The duration between record produce time at source and replicated to target. Maximum Milliseconds
Message Lag Indicates the number of messages the replicator is behind the source cluster. Maximum Count
Replicator Throughput The average number of bytes replicated per second. Average Bytes
Replicator Failure The number of failures that the replicator is experiencing. Sum Count

MSK Connect

The supported metrics for MSK Connect monitor are provided below:

Metric name Description Statistics Unit
Bytes In Per Sec The total number of bytes received by the connector. Average Bytes
Bytes Out Per Sec The total number of bytes delivered by the connector. Average Bytes
CPU Utilization The percentage of CPU consumption by system and user. Average Percentage
Errored Task Count The number of tasks that have errored out. Sum Count
Memory Utilization The percentage of the total memory on a worker instance. Maximum Percentage
Rebalance Completed Total The total number of rebalances completed by this connector. Maximum Count
Rebalance Time Average The average time in milliseconds spent by the connector on rebalancing. Average Milliseconds
Rebalance Time Maximum The maximum time in milliseconds spent by the connector on rebalancing. Maximum Milliseconds
Rebalance Time Since Last The time in milliseconds since this connector completed the most recent rebalance. Average Milliseconds
Running Task Count The running number of tasks in the connector. Maximum Count
Sink Record Read Rate The average per-second number of records read from the Apache Kafka or Amazon MSK Cluster. Average Count
Sink Record Send Rate The average per-second number of records that are output from the transformations and sent to the destination.  Average Count
Source Record Poll Rate The average per-second number of records produced or polled. Average Count
Source Record Write Rate The average per-second number of records output from the transformations and written to the Apache Kafka or Amazon MSK Cluster. Average Count
Task Startup Attempts Total The total number of task startups that the connector has attempted.  Sum Count
Task Startup Success Percentage The average percentage of successful task starts for the connector.  Average Percentage
Worker Count The number of workers that are running in the connector. Maximum Count

MSK Broker

The supported metrics for MSK Broker monitor are provided below:

Metric name Description Statistics Unit
CPU Credit Balance The number of earned CPU credit used by a broker. Maximum Count
Burst Balance The credit balance available for the EBS Volume monitors in the cluster. Sum Count
Client Connection Count The number of active authenticated client connections. Maximum Count
Connection Count The number of active authenticated, unauthenticated, and inter-broker connections. Maximum Count
CPU Idle The percentage of CPU idle time. Average Percentage
CPU IO Wait The percentage of CPU idle time during a pending disk operation. Average Percentage
CPU System The percentage of CPU that has kernel space. Average Percentage
CPU User The percentage of CPU that has user space. Average Percentage
Kafka AppLogs Disk Used The maximum disk space used for application logs. Maximum Percentage
Kafka DataLogs Disk Used The maximum disk space used for data logs. Maximum Percentage
Leader Count The number of leaders for the partitions without including the replicas. Maximum Count
Memory Buffered The size in bytes of buffered memory for the broker. Average Bytes
Memory Cached The size in bytes of cached memory for the broker. Average Bytes
Memory Free The size in bytes of memory that is free and available for the broker. Average Bytes
Heap Memory After GC The percentage of total heap memory in use after garbage collection. Average Percentage
Memory Used The size in bytes of the memory that is in use for the broker. Average Bytes
Messages In Per Sec The number of messages received per second. Maximum Count
Network Rx Dropped The number of dropped receive packages. Sum Count
Network Rx Errors The number of network receive errors for the broker. Sum Count
Network Rx Packets The number of packets received by the broker. Sum Count
Network Tx Dropped The number of dropped transmit packages. Sum Count
Network Tx Errors The number of network transmit errors for the broker. Sum Count
Network Tx Packets The number of packets transmitted by the broker. Sum Count
Offline Partitions Count The total number of partitions that are offline in the cluster. Maximum Count
Partition Count The number of partitions per broker, including replicas. Maximum Count
Produce Total Time Ms Mean The mean produce time in milliseconds. Average Milliseconds
Request Bytes Mean The mean number of request bytes for the broker. Average Bytes
Request Time The average time in milliseconds spent in the broker network and I/O threads to process requests. Average Milliseconds
Root Disk Used The percentage of the root disk used by the broker. Maximum Percentage
Swap Free The size in bytes of swap memory that is available for the broker. Average Bytes
Swap Used The size in bytes of swap memory that is in use for the broker. Average Bytes
Traffic Shaping The number of packets shaped (dropped or queued) due to exceeding network allocations. Maximum Count
Under MinIsr Partition Count The number of under minIsr partitions for the broker. Maximum Count
Under Replicated Partitions The number of under-replicated partitions for the broker. Maximum Count
ZooKeeper Request Latency Ms Mean The mean latency for Apache ZooKeeper requests from the broker. Average Milliseconds
ZooKeeper Session State The connection status of the broker's ZooKeeper session which may be one of the following: NOT_CONNECTED: '0.0', ASSOCIATING: '0.1', CONNECTING: '0.5', CONNECTEDREADONLY: '0.8', CONNECTED: '1.0', CLOSED: '5.0', AUTH_FAILED: '10.0'. Maximum Count

Per Broker

If your monitoring level is set to PER_BROKER in AWS, you get the metrics described in the following table in addition to all the DEFAULT level metrics. The PER_BROKER metrics are available under the Advanced tab.

The supported metrics for Per Broker monitor are provided below:

Metric name Description Statistics Unit
Bandwidth In Allowance Exceeded The number of packets shaped because the inbound aggregate bandwidth exceeded the maximum for the broker. Maximum Count
Bandwidth Out Allowance Exceeded The number of packets shaped because the outbound aggregate bandwidth exceeded the maximum for the broker. Maximum Count
Connection Track Allowance Exceeded The number of packets shaped because the connection tracking exceeded the maximum for the broker.  Maximum Count
Connection Close Rate The number of connections closed per second per listener.  Sum Count
Connection Creation Rate The number of new connections established per second per listener. Sum Count
CPU Credit Usage The number of CPU credits spent by the broker. Sum Count
Fetch Consumer Local Time Ms Mean The mean time in milliseconds that the consumer request is processed at the leader. Average Milliseconds
Fetch Consumer Request Queue Time Ms Mean The mean time in milliseconds that the consumer request waits in the request queue. Average Milliseconds
Fetch Consumer Response Queue Time Ms Mean
The mean time in milliseconds that the consumer request waits in the response queue.
Average Milliseconds
Fetch Consumer Response SendTime Ms Mean
The mean time in milliseconds for the consumer to send a response.
Average Milliseconds
Fetch Consumer TotalTime Ms Mean The mean total time in milliseconds that consumers spend on fetching data from the broker. Average Milliseconds
Fetch Follower LocalTime Ms Mean The mean time in milliseconds that the follower request is processed at the leader. Average Milliseconds
Fetch Follower Request Queue Time Ms Mean The mean time in milliseconds that the follower request waits in the request queue. Average Milliseconds
Fetch Follower Response QueueTime Ms Mean The mean time in milliseconds that the follower request waits in the response queue. Average Milliseconds
Fetch Follower Response SendTime Ms Mean The mean time in milliseconds for the follower to send a response. Average Milliseconds
Fetch Follower TotalTime Ms Mean The mean total time in milliseconds that followers spend on fetching data from the broker. Average Milliseconds
Fetch Message Conversions Per Sec The number of fetch message conversions per second for the broker. Average Count
Fetch Throttle Byte Rate The number of throttled bytes per second. Average Bytes
Fetch Throttle Queue Size The number of messages in the throttle queue. Average Count
Fetch Throttle Time The average fetch throttle time in milliseconds. Average Milliseconds
Network Processor Average Idle Percent The average percentage of the time the network processors are idle. Average Percentage
PPS Allowance Exceeded The number of packets shaped because the bidirectional PPS exceeded the maximum for the broker. Maximum Count
Produce LocalTime Ms Mean The mean time in milliseconds that the request is processed at the leader. Average Count
Produce Message Conversions Per Sec The number of produce message conversions per second for the broker. Average Count
Produce Message Conversions Time Ms Mean The mean time in milliseconds spent on message format conversions. Average Milliseconds
Produce Request Queue Time Ms Mean The mean time in milliseconds that request messages spend in the queue. Average Milliseconds
Produce Response Queue Time Ms Mean The mean time in milliseconds that response messages spend in the queue. Average Milliseconds
Produce Response SendTime Ms Mean The mean time in milliseconds spent on sending response messages. Average Milliseconds
Produce Throttle Byte Rate The number of throttled bytes per second. Maximum Bytes
Produce Throttle Queue Size The number of messages in the throttle queue. Maximum Count
Produce Throttle Time The average produce throttle time in milliseconds. Average Milliseconds
Remote Fetch Bytes Per Sec The total number of bytes transferred from tiered storage in response to consumer fetches. Maximum Bytes
Remote Copy Bytes Per Sec The total number of bytes transferred to tiered storage, including data from log segments, indexes, and other auxiliary files. Maximum Bytes
Remote LogManager Tasks Average Idle Percent The average percentage of time the remote log manager spent idle. Average Percentage
Remote LogReader Average Idle Percent The average percentage of time the remote log reader spent idle. Average Percentage
Remote LogReader Task Queue Size The number of tasks responsible for reads from tiered storage that are waiting to be scheduled. Sum Count
Remote Fetch Errors Per Sec The total rate of errors in response to read requests that the specified broker sent to tiered storage to retrieve data in response to consumer fetches. Average Count
Remote Fetch Requests Per Sec The total number of read requests that the specifies broker sent to tiered storage to retrieve data in response to consumer fetches. Average Count
Remote Copy Errors Per Sec The total rate of errors in response to write requests that the specified broker sent to tiered storage to transfer data upstream. Average Count
Replication Bytes In Per Sec The number of bytes per second received from other brokers. Average Bytes
Replication Bytes Out Per Sec The number of bytes per second sent to other brokers. Average Bytes
Request Exempt From Throttle Time The average time in milliseconds spent in the broker network and I/O threads to process requests that are exempt from throttling. Average Milliseconds
Request Handler Average Idle Percent The average percentage of the time the request handler threads are idle. Average Percentage
Request Throttle Queue Size The number of messages in the throttle queue. Average Count
Request Throttle Time The average request throttle time in milliseconds. Average Milliseconds
TCP Connections Shows number of incoming and outgoing TCP segments with the Synchronization (SYN) flag set. Sum Count
Remote Copy Lag Bytes The total number of bytes of the data that is eligible for tiering on the broker but has not been transferred to tiered storage yet. Sum Bytes
Traffic Bytes Shows network traffic in overall bytes between clients (producers and consumers) and brokers. Maximum Bytes
Volume Queue Length The number of read and write operation requests waiting to be completed in a specified time period. Maximum Count
Volume Read Bytes The number of bytes read in a specified time period. Sum Bytes
Volume Read Ops The number of read operations in a specified time period. Sum Count
Volume Total ReadTime The total number of seconds spent by all read operations that completed in a specified time period. Sum Seconds
Volume Total WriteTime The total number of seconds spent by all write operations that completed in a specified time period. Sum Seconds
VolumeWriteBytes The number of bytes written in a specified time period. Sum Bytes
Volume Write Ops The number of write operations in a specified time period. Sum Count

MSK Topic

The supported metrics for MSK Topic monitor are provided below:

Metric name Description Statistics Unit
Bytes In Per Sec The number of bytes per second received from clients. Average Bytes
Bytes Out Per Sec The number of bytes per second sent to clients.  Average  Bytes
Fetch Message Conversions Per Sec The number of fetch message conversions. Average Count
Messages In Per Sec The number of incoming messages per second for the topic. Average Count
Produce Message Conversions Per Sec The number of produce message conversions per second. Average Count
Remote Fetch Bytes Per Sec The number of bytes transferred from tiered storage in response to consumer fetches for the specified topic and broker. Average Bytes
Remote Copy Bytes Per Sec The number of bytes transferred to tiered storage, for the specified topic and broker. Average Bytes
Remote Fetch Errors Per Sec The rate of errors in response to read requests that the specified broker sends to tiered storage to retrieve data in response to consumer fetches on the specified topic. Maximum Count
Remote Fetch Requests Per Sec The number of read requests that the specified broker sends to tiered storage to retrieve data in response to consumer fetches on the specified topic. Maximum Count
Remote Copy Errors Per Sec The rate of errors in response to write requests that the specified broker sends to tiered storage to transfer data upstream. Maximum Count
Total Max Offset Lag The maximum offset lag across all partitions in a topic. Maximum Count
Total Estimated Max Time Lag

The estimated time it would take for the consumer to catch up to the latest message, considering the biggest gap (MaxOffsetLag).

If this number is high, it means there is a significant delay before messages are read, indicating that the consumer is lagging behind and may need adjustments to keep up with the message production rate.

Maximum Milliseconds
Total Sum Offset Lag The aggregated offset lag for all the partitions in a topic. Sum Count
Total Rolling Estimated Time Lag Max The highest estimated time lag averaged over a rolling time window. This provides a stable view of the maximum delay in message consumption over a period of time, helping you understand if there are consistent delays in message processing. Maximum Milliseconds

Threshold configuration

To configure thresholds for Amazon MSK:

  1. Log in to your Site24x7 account and navigate to Admin > Configuration Profiles > Threshold and Availability.
  2. Click Add Threshold Profile.
  3. Select the applicable monitor type from the Monitor Type drop-down menu and provide an appropriate name in the Display Name field. The applicable monitor types are MSK Cluster, MSK Serverless, MSK Replicator, MSK Connect, MSK Broker, and MSK Topic.
  4. The supported metrics are displayed in the Threshold Configuration section. You can set threshold values for all the metrics mentioned above.
  5. Click Save.
Note

You can manage all the alerts to ensure that the notifications are sent through the parent monitor with the status propagation feature.

Licensing

  • MSK Cluster: Each MSK Cluster monitor is considered a basic monitor.
  • MSK Serverless: The MSK Serverless monitor is free.
  • MSK Replicator: Each MSK Replicator monitor is considered a basic monitor.
  • MSK Connect: Each MSK Connect monitor is considered a basic monitor.
  • MSK Broker: Each MSK Broker monitor is considered a basic monitor.
  • MSK Topic: Each MSK Topic monitor is considered a basic monitor.

Viewing Amazon MSK data

To monitor your Amazon MSK, log in to your Site24x7 account and navigate to Cloud > AWS > Amazon MSK.

Amazon MSK monitor data

The monitor data for each Amazon MSK monitor is given below.

MSK Cluster

You can view your MSK Cluster data on the following tabs.

Summary

The Summary tab provides an overview of the events timeline and metrics in the form of charts.

Broker

You can view the Broker Availability along with the Broker monitors associated with your cluster in this tab. Click the applicable Broker monitor from the Broker list to view the monitor details.

Topic

The Topic tab displays the Topic Availability and Topic instances associated with your cluster. Click the applicable Topic instance from the Topic list to view the monitor details.

Configuration 

The Configuration tab summarizes essential details of your MSK cluster, such as Cluster ARN, Cluster Name, Current State, and Number of Brokers.

Outages

The Outages tab provides details on an outage's start time, end time, duration, and comments (if any).

Inventory

Obtain details like Cluster Name, Region, and Monitor Licensing Category from the Inventory tab. Set and view the Threshold and Availability Profile and the Notification Profile according to the user in this tab.

Log Report

This tab offers a consolidated report of the MSK Cluster monitor's log status, which can be downloaded as a CSV file.

MSK Serverless

You can view your MSK Serverless data on the following tabs.

Summary

The Summary tab provides an overview of the events timeline and metrics in the form of charts.

Topic

The Topic tab displays the Topic Availability and Topic instances associated with your cluster. Click the applicable Topic instance from the Topic list to view the monitor details.

Configuration

The Configuration tab summarizes essential details of your MSK Serverless, such as Cluster ARN, Cluster Name, Current State, and Created Date.

Outages

The Outages tab provides details on an outage's start time, end time, duration, and comments (if any).

Inventory

Obtain details like Cluster Name, Region, and Monitor Licensing Category from the Inventory tab. Set and view the Threshold and Availability Profile and the Notification Profile according to the user in this tab.

Log Report

This tab offers a consolidated report of the MSK Serverless monitor's log status, which can be downloaded as a CSV file.

MSK Replicator

You can view your MSK Replicator data under the following tabs.

Summary

The Summary tab provides an overview of the events timeline and metrics in the form of charts.

Monitored Resources

The Monitored Resources tab in the MSK Replicator monitor provides the monitored resource details of the MSK clusters represented by the MSK Replicator monitor. Click the monitor display name to view the MSK monitor details.

Configuration 

The Configuration tab summarizes essential details of your MSK Replicator, such as Replicator Name, Replicator ARN, Replicator State, and Replicator Version.

Outages

The Outages tab provides details on an outage's start time, end time, duration, and comments (if any).

Inventory

Obtain details like Replicator ID, Region, and Monitor Licensing Category from the Inventory tab. Set and view the Threshold and Availability Profile and the Notification Profile according to the user in this tab.

Log Report

This tab offers a consolidated report of the MSK Replicator monitor's log status, which can be downloaded as a CSV file.

MSK Connect

You can view your MSK Connect data on the following tabs.

Summary

The Summary tab provides an overview of the events timeline and metrics in the form of charts.

Outages

The Outages tab provides details on an outage's start time, end time, duration, and comments (if any).

Configuration

The Configuration tab summarizes essential details of your MSK cluster such as Connect Name, Connect ARN, Service Role ARN, and Current Version.

Inventory

Obtain details like Connect Name, Region, and Monitor Licensing Category from the Inventory tab. Set and view the Threshold and Availability Profile and the Notification Profile according to the user in this tab.

Log Report

This tab offers a consolidated report of the MSK Connect monitor's log status, which can be downloaded as a CSV file.

MSK Broker

You can view your MSK Broker data on the following tabs.

Summary

The Summary tab provides an overview of the events timeline and metrics in the form of charts.

Advanced

Site24x7 offers an advanced metrics tab for MSK Broker. The Advanced tab provides crucial insights into various metrics such as Network Traffic, CPU Utilization, Request Handling, and Replication, and more. This tab enables users to effectively monitor and optimize the performance of your MSK Broker.

Note

 

The Advanced tab will be displayed only if you have set the monitoring level to PER_BROKER for your AWS account.

Topic

The Topic tab displays the Topic Availability and Topic instances associated with your cluster. Click the applicable Topic instance from the Topic list to view the monitor details.

Outages

The Outages tab provides details on an outage's start time, end time, duration, and comments (if any).

Inventory

Obtain details like Cluster Name, Broker ID, Region, and Monitor Licensing Category from the Inventory tab. Set and view the Threshold and Availability Profile and the Notification Profile in this tab.

Log Report

This tab offers a consolidated report of the MSK Broker monitor's log status, which can be downloaded as a CSV file.

MSK Topic

You can view your MSK Topic data on the following tabs.

Summary

The Summary tab offers a comprehensive overview of the events timeline and metrics, presenting insightful charts that shed light on the performance and behavior of MSK broker instances and topics within your Kafka environment.

Outages 

The Outages tab provides details on an outage's start time, end time, duration, and comments (if any).

Inventory

Obtain details like Broker ID, Topic Name, Region, and Monitor Licensing Category from the Inventory tab. Set and view the Threshold and Availability Profile and the Notification Profile according to the user in this tab.

Log Report

This tab offers a consolidated report of the MSK Topic monitor's log status, which can be downloaded as a CSV file.

Was this document helpful?

Would you like to help us improve our documents? Tell us what you think we could do better.


We're sorry to hear that you're not satisfied with the document. We'd love to learn what we could do to improve the experience.


Thanks for taking the time to share your feedback. We'll use your feedback to improve our online help resources.

Shortlink has been copied!