Validation
How Telflo checks a configuration as you build it.
Telflo validates a configuration continuously, so problems surface before you test or deploy. Validation runs at two levels: structural checks on the canvas graph, and schema checks against the collector binary for the selected version.
Structural checks
Structural checks cover the connection and graph rules.
| Rule | Severity |
|---|---|
| A receiver has an outgoing connection | Warning |
| An exporter has an incoming connection | Warning |
| A processor has incoming and outgoing connections | Warning |
| A connection respects the component types | Error |
| A connector joins two pipelines of compatible signal types | Error |
| An extension has no connections | Error |
Connections that break the component-type rules are also prevented as you draw them.
Schema checks
Telflo validates the configuration against the real collector binary for the collector version selected in the toolbar. Unknown components, unknown fields, and invalid values are reported.
Schema validation also reports deployment requirements: values the collector needs from its environment, such as host paths or environment variables. Requirements do not block saving. They state what the deploy environment must provide.
Configuration providers
Only ${env:VAR} references are evaluated during validation. References to other configuration providers, such as ${file:...}, ${http:...}, ${https:...}, and ${yaml:...}, are replaced with placeholders before the collector validates, and the substitution is reported as a deployment requirement. The referenced files and URLs are read in the real deployment, not during validation.
The same restriction applies in the testing sandbox: non-env provider references are replaced with a placeholder before a test runs, so a test cannot read files or the network from the test environment. The substitution is listed with the run's sandbox mutations.
Validation status
The YAML view shows a status row with the result: Config valid, or the number of errors. Expand it to see each reported issue, with the affected component and a description, and the configuration's deployment requirements.
While you edit YAML manually, the canvas shows Yaml is Valid when the edits pass, or Errors must be corrected. when they do not, and the edits cannot be applied to the canvas until every error is resolved.
Validation and testing
Validation answers whether a configuration is well-formed and valid. To check what a configuration does to real data, run a test. See Testing.
Last updated on