Skip to content

Latest commit

 

History

History
83 lines (68 loc) · 3.88 KB

configuration.md

File metadata and controls

83 lines (68 loc) · 3.88 KB
Raw

Configuration

Slo exporter itself is configured using one base YAML file. Path to this file is configured using the --config-file flag. Additional configuration files might be needed by some modules depending on their needs and if they are used in the pipeline at all.

ENV variables

Every configuration option in the base YAML file can be overridden (the configuration MUST be present in the yaml config for the ENV to be loaded, it can be even empty string) by using ENV variable. The schema of ENV variable naming is SLO_EXPORTER_ prefix and than in uppercase any key of the YAML structure in uppercase without any underscores. Underscores are used for separating nested structures. Example: SLO_EXPORTER_WEBSERVERLISTENADDRESS=0.0.0.0:8080 or for module configuration SLO_EXPORTER_TAILER_TAILEDFILE=access.log

CMD flags

$ ./slo_exporter --help-long
usage: slo_exporter --config-file=CONFIG-FILE [<flags>]

Flags:
  --help                     Show context-sensitive help (also try --help-long and --help-man).
  --config-file=CONFIG-FILE  SLO exporter configuration file.
  --log-level="info"         Log level (error, warn, info, debug,trace).
  --check-config             Only check config file and exit with 0 if ok and other status code if not.

Processing pipeline

Slo-exporter allows to dynamically compose the pipeline structure, but there is few basic rules it needs to follow:

  • ingester module cannot be at the beginning of pipeline.
  • ingester module can only be linked to preceding producer module.
  • Type of produced event by the preceding module must match the ingested type of the following one.

Base config

# Address where the web interface should listen on.
webServerListenAddress: "0.0.0.0:8080"
# Maximum time to wait for all events to be processed after receiving SIGTERM or SIGINT.
maximumGracefulShutdownDuration: "10s"
# How long to wait after processing pipeline has been shutdown before stopping http server w metric serving.
# Useful to make sure metrics are scraped by Prometheus. Ideally set it to Prometheus scrape interval + 1s or more.
# Should be less or equal to afterPipelineShutdownDelay
afterPipelineShutdownDelay: "1s"

# Defines architecture of the pipeline how the event will be processed by the modules.
pipeline: [<moduleType>]

# Contains configuration for distinct pipeline module.
modules:
  <moduleType>: <moduleConfig>

moduleType:

Producers:

Only produces new events from the specified data source.

Processors:

Reads input events, does some processing based in the module type and produces modified event.

Ingesters:

Only reads input events but does not produce any.

Details how they work and their moduleConfig can be found in their own linked documentation in the docs/modules folder.

Configuration examples

Actual examples of usage with full configuration can be found in the examples/ directory.

Configuration testing

If you want to verify that your configuration is valid, use the --check-config flag. Slo-exporter then just verifies if the configuration is valid and exits with status 0 if ok and 1 if not.