vector(1)
Vector is a lightweight, open-source data pipeline tool for collecting, transforming, and routing logs and metrics.
Synopsis
vector [COMMAND] [OPTIONS]Description
Vector is a high-performance, end-to-end (agent and aggregator) platform for building observability pipelines. It collects logs, metrics, and traces from various sources, transforms them using a simple configuration language, and routes them to multiple destinations.
Vector runs as a standalone daemon or can be embedded in applications. It supports dozens of input sources (files, syslog, Kafka, HTTP) and output sinks (Datadog, Elasticsearch, CloudWatch, S3, and many others), making it ideal for centralizing observability data without vendor lock-in.
Common options
| Flag | What it does |
|---|---|
-c, --config FILE | Read configuration from specified file (TOML, YAML, or JSON) |
-d, --dry-run | Validate configuration without starting the daemon |
-q, --quiet | Suppress informational output (only show warnings and errors) |
-vv | Enable debug logging (twice for extra verbosity) |
--watch-config | Automatically reload configuration when files change |
-l, --log-level LEVEL | Set logging level (trace, debug, info, warn, error) |
--version | Display Vector version and exit |
--help | Show help message and available commands |
Examples
Check installed Vector version
vector --versionStart Vector daemon with configuration from /etc/vector/vector.toml
vector -c /etc/vector/vector.tomlValidate configuration file without actually running Vector
vector -d -c config.tomlRun Vector with debug logging and automatically reload when config files change
vector -c config.toml --watch-config -vvValidate the configuration file and show any errors or warnings
vector validate config.tomlShow real-time metrics about Vector's performance and data throughput
vector topRun Vector silently, only showing errors and warnings
vector -c config.toml --quiet