randomfailureprocessor

Random Failure Processor

The random failure processor is used for testing error resiliency in the collector.

Supported pipelines

• Logs
• Metrics
• Traces

How it works

1. The user configures the processor in one or more pipelines.
2. Whenever telemetry passes through the processor, there is a user-configured probability that an error will be returned.

Configuration

Field	Type	Default	Required	Description
failure_rate	float64	0.5	false	The probability, between 0 and 1, of any given piece of telemetry causing an error.
error_message	string	"random failure"	false	The error message that will be returned by the processor.

Examples

Usage in pipelines

The random failure processor may be used in a pipeline in order to test what occurs when a section of the pipeline errors:

receivers:
  file_log:
    include: [/var/log/logfile.txt]

processors:
  randomfailure:
    failure_rate: 0.1
    error_message: "10% failure occurred!"

exporters:
  nop:

service:
  pipelines:
    logs:
      receivers: [file_log]
      processors: [randomfailure]
      exporters: [nop]

In this instance, each log from /var/log/logfile.txt has a 10% chance of generating an error when it passes through the processor.