Monitoring Amazon Timestream

Instana comprehensively monitors your Amazon Timestream databases and provides visibility into your environment by using the Amazon Timestream sensor. After you install the Instana agent, the Amazon Timestream sensor is automatically deployed and installed. You can view metrics that are related to Amazon Timestream databases in the Instana UI.

Currently, monitoring of Amazon Timestream for only LiveAnalytics is available. Monitoring of InfluxDB instances is not supported.

Amazon Timestream for LiveAnalytics is a fast, scalable, fully-managed, purpose-built time series database that stores and analyzes trillions of time series data points per day. For more information about other supported AWS services, see the AWS topic.

Required IAM permissions

The Amazon Timestream sensor requires the following AWS IAM permissions to monitor your Amazon Timestream databases:

  • cloudwatch:GetMetricData
  • cloudwatch:GetMetricStatistics
  • timestream:ListDatabases
  • timestream:DescribeEndpoints
  • timestream:DescribeDatabase
  • timestream:ListTagsForResource

Configuring the Amazon Timestream sensor

After the Instana agent is installed, the Amazon Timestream sensor is enabled by default. Metrics for Amazon Timestream databases are pulled every 60 seconds. You can change this setting in the <agent_install_dir>/etc/instana/configuration.yml file:

com.instana.plugin.aws.timestream:
  cloudwatch_period: 60

Enabling proxy configuration

To enable the Amazon Timestream sensor to use the proxy configuration, add the following agent configuration settings:

com.instana.plugin.aws.timestream:
  proxy_host: 'example.com' # proxy host name or ip address
  proxy_port: 3128 # proxy port
  proxy_protocol: 'HTTP' # proxy protocol: HTTP or HTTPS
  proxy_username: 'username' # OPTIONAL: proxy username
  proxy_password: 'password' # OPTIONAL: proxy password

Disabling the Amazon Timestream sensor

To disable monitoring of the Amazon Timestream databases, use the following configuration:

com.instana.plugin.aws.timestream:
  enabled: false

Configuring monitoring for Amazon Timestream across multiple AWS accounts

To set up monitoring of multiple AWS accounts with one AWS agent in the same region, you can use two approaches. For more information, see Monitoring multiple AWS accounts.

AWS-named profiles approach

To override the profiles that monitor Amazon Timestream, use the following configuration:

com.instana.plugin.aws.timestream:
  profile_names:
    - 'profile2'
    - 'profile3'

If you define profiles on a service level, the global AWS profile configuration is overridden.

AWS STS approach

To override IAM roles that monitor the Amazon Timestream, use the following configuration:

com.instana.plugin.aws.timestream:
  role_arns:
    - 'arn:aws:iam::<account_1_id>:role/<role_1_name>'
    - 'arn:aws:iam::<account_2_id>:role/<role_2_name>'

If you define IAM roles on a service level, the global AWS IAM role configuration is overridden.

Filtering by defining tags

Instana monitors all Amazon Timestream databases by default. You can also filter which Amazon Timestream databases Instana monitors by defining tags in the configuration.yaml file. If you don't want to filter databases, you needn't define the tags.

To define multiple tags, separate them with commas. Define tags as a key-value pair. To simplify configuration, you can define which tags you want to include in discovery or exclude from discovery. If you define a tag in both lists (include and exclude), the exclude list take precedence. It’s not mandatory to define all values to enable filtering.

You can specify how often sensors poll the AWS tagged resources by using the tagged-services-poll-rate configuration property. By default, the value is 300 seconds.

Tags are available with only the AWS agent.

To define how often sensors poll the tagged resources, use the following configuration:

com.instana.plugin.aws:
  tagged-services-poll-rate: 60 #default 300

To include Amazon Timestream databases with tags into discovery, use the following configuration:

com.instana.plugin.aws.timestream:
  include_tags: # Comma separated list of tags in key:value format (e.g. env:prod,env:staging)

To exclude Amazon Timestream databases with tags from discovery, use the following configuration:

com.instana.plugin.aws.timestream:
  exclude_tags: # Comma separated list of tags in key:value format (e.g. env:dev,env:test)

AWS services without tags are monitored by default, but can be excluded by setting the include_untagged field to false:

com.instana.plugin.aws.timestream:
  include_untagged: false # True value by default

Viewing metrics

To view the metrics, complete the following steps:

  1. From the navigation menu in the Instana UI, select Infrastructure.
  2. Click a specific monitored Amazon Timestream database.

You can see the Amazon Timestream database dashboard with all the collected metrics.

Configuration data

The Amazon Timestream sensor collects the following configuration data:

Name Description
ARN The Amazon Resource Name (ARN) of the Timestream database.
Database name The name of the Timestream database.
Created time The date and time of creation of the database.
Type LiveAnalytics is available. Monitoring of InfluxDB instances is not supported.

Metrics

The Amazon Timestream sensor collects the following metrics:

Name Description
Successful Request Latency The latency of successful requests to Timestream for LiveAnalytics during the specified time period. This metric is available for the following statistics: Minimum, Maximum, Average, and SampleCount.
Magnetic Store Rejected Record Count The number of magnetic store-written records that are rejected asynchronously.
Active Magnetic Store Partitions The number of magnetic store partitions that actively ingest data at a particular time.
Magnetic Store Pending Records Latency The oldest write to a magnetic store that is not available for query.
Magnetic Store Rejected Upload User Failures The number of magnetic store-rejected record reports that are not uploaded due to user errors.
Magnetic Store Rejected Upload System Failures The number of magnetic store rejected record reports that are not uploaded due to system errors.