Edge Delta Global Settings

Configure Edge Delta Data Pipeline Preferences.


Edge Delta Fleets send certain data to the Edge Delta back end, for example agent health data and system events. In the Edge Delta App, click Pipelines, and then click Settings.

Agent Self Logs

This data type represents the output of agent logs, which helps Edge Delta Support and engineering to troubleshooting agent behavior.


This data type represents a signal detected by the agent regarding an unexpected event. This information and corresponding metadata are sent to the Edge Delta backend by default. This data type is based on the threshold settings in the Pipeline configuration.

Log Archiving

This data type represents the compressed storage of raw logs on a cloud storage service, such as S3. Edge Delta maintains separate S3 buckets per customer organization. This data type populates the Log Search page where you can filter for specific logs.

Due to an ingestion delay of archived files, logs in from the past few minutes may not display.

Log Patterns

Patterns are reduced versions of the original log line where specific parts are placed with asterisks *. Typically, these patterns do not contain sensitive data. This data type can be enabled by adding a cluster processor.

Log Pattern Samples

This data type represents the raw logs that match a specific pattern. For each distinct cluster pattern found by the agent, the agent also sends a few samples of that cluster.

Alert Contextual Logs

This data type represents the logs generated from an issue. An issue can be an anomaly detected by an individual agent or by the backend processing jobs. Contextual logs are useful to investigate an issue and find a root cause. In the Edge Delta App, you can use the Metrics - Anomalies page to view all signals and findings.


An event is a Kubernetes-based anomaly. By default, Edge Delta consumes Kubernetes system events, and then displays those events on the Metrics - Anomalies page.

Agent Health Data

This data type represents the various critical internal components of the agent. This data includes:

  • Success count
  • Failure count
  • Last error
  • Start time

Agent Heartbeats

This data type represents the heartbeat that the agent sends to the backend every minute. Edge Delta uses this information to ensure that the agent is running. This information populates the Pipeline Status page of the Edge Delta App.


This data type can be collected from the environment, such as system_stats, or generated by a processor.

Turning off metrics will disable the pipeline status page, which is populated by metrics.

To change data ingestion preferences you need to contact Edge Delta.

You can also set automatic monitor creation. This configures a set of default monitors upon account creation.

Configuration Variables

If your account contains multiple Fleets, you may want to consider setting variables that can be globally applied. These variables can be referenced in you Pipeline configurations either with or without a fallback default value:

  • '{{ Env "MY_VARIABLE_NAME" }}'
  • '{{ Env "MY_VARIABLE_NAME" "my default value" }}'

If the default value is not provided and there is no variable set the Agent will stop with an error. If a default value is provided, but the variable does not exist, the default value will be used.

Environment variables can only be used with string-typed inputs.

  - name: my_http_port
	type: http_output
	endpoint: '{{ Env "MY_HTTP_ENDPOINT" }}'

Sentiment Analysis

Every pattern detected by the Edge Delta agent is analyzed to check for negative sentiment. Negative sentiment is determined by checking for the presence of specific keywords or Golang Regex patterns. Some keywords such as debug are considered neutralizing because they automatically offset negative keyword matches in the pattern. Sentiment keywords are applied to all Fleets within the account.

You can specify Golang Regex such as alert([=\s\*]+|$) or string literals such as broken. Literals are not case sensitive.