Releases: julianstephens/coolify-deploy
Releases · julianstephens/coolify-deploy
Release list
v0.2.6
- Fixed environment variable parsing for
.envsecrets that include whitespace around the=sign, e.g.KEY = value, which previously caused valid variables to be dropped before reaching Coolify. - Fixed
COOLIFY_ENV_*secret generation and consumption to properly handle names with spaces, punctuation, or path-like values while preserving the expected naming convention. - Added support for the
--server-uuidalias while keeping--server-idas the canonical CLI option, improving compatibility with docs and real-world usage. - Fixed manifest-level env fallback handling so a configured
envFileSecretNameis respected when resource-specific secrets are absent.
v0.2.5
- Added User-Agent header to API requests to prevent rejection by API servers that filter automated requests
- Added debug logging for token validation (logs token length and preview without exposing full token) to help diagnose authentication issues between environments
v0.2.4
- Fixed bug where API token was not being trimmed of whitespace, causing authentication errors. The tool now automatically trims the token before using it in API requests, preventing common issues with copy-pasting tokens that include extra spaces.
v0.2.3
- Fixed bug where
custom_healthcheck_foundwas still in use during app update. - Fixed bug where apps without a network config failed creation due to missing
portsExposesfield
v0.2.2
- Aligned with Coolify API changes by removing the deprecated custom_healthcheck_found field from application update payloads. This resolves validation errors when updating
- Fixed a bug where creating new applications would fail due to a missing ports_exposes field. The tool now correctly includes this field in the creation payload.Fixed a bug where creating new applications would fail due to a missing ports_exposes field. The tool now correctly includes this field in the creation payload.
- Corrected the application creation logic to use the portsExposes property from the manifest, ensuring consistent port configuration between created and updated applications.
v0.2.1
- Forces
custom_healthcheck_foundtotruewhen updating health check configuration. This ensures that Coolify respects the custom health check settings provided in the manifest. - Improves error logging for API requests. Now logs the full response body from the Coolify API when a request fails, providing more details on validation errors.
- Adds debug logging for application creation payloads. This helps troubleshoot validation errors when creating new applications, specifically for cases where required fields might be missing or invalid.
v0.2.0
- Enforces strict validation on the manifest file. This ensures that typos in configuration keys (like
healthcheckinstead ofhealthCheck) are caught immediately instead of being silently ignored. - Adds debug logging for application update payloads to help troubleshoot configuration issues.
v0.1.8
- Optimizes deployment polling to run concurrently. Instead of waiting for each application to finish deploying one by one, the tool now triggers all deployments first and then waits for them in parallel, significantly reducing total runtime.
- Fixes "Validation failed" error when creating or updating applications with empty
portsExposesin the manifest. The tool now correctly omits the field instead of sending an empty string.
v0.1.7
- Improves error messages for API permission errors (401/403) to explicitly suggest checking token scopes.
- Fixes deployment trigger error "You can only use uuid or tag, not both" by removing the redundant
tagparameter when deploying byuuid.
v0.1.6
- Extends types to match API documentation
- Fixes issue where internal API client crashed on successful but empty response
- Fixes issue where deployments were not triggered after creating or updating
- Fixes issue where deployments accidentally target the wrong application if multiple apps share the same name across different environments (e.g., "api" in both Staging and Production)
- Fixes issue where dry-run mode failed with "Target environment does not exist" error.
- Improves dry-run mode to perform real read-only API calls (list applications, environments, etc.) instead of returning mock data. This allows the dry-run to accurately predict whether resources will be created or updated based on the actual server state. In addition, the tool now prints a summary table of changes (Created, Updated, Pruned, Failed) when running in dry-run mode, acting as a "plan" or "drift detection" feature.
- Adds deployment status polling. The tool now waits for deployments to finish (or fail) before exiting, ensuring that CI pipelines accurately reflect the deployment outcome.
- Adds resource pruning. Applications that exist in the target environment but are not defined in the manifest will now be deleted. This ensures the environment exactly matches the manifest.