ingestr server

The server command starts the ingestr web UI. The UI lets you save reusable connections, configure an ingestion run, stream logs while the run is active, and review previous runs.

ingestr server

By default, the server starts on port 8080 and prints the URL:

Starting ingestr web UI on http://localhost:8080

Open that URL in your browser to use the UI.

Options

Flag	Environment variable	Default	Description
--port	INGESTR_PORT	8080	Port to listen on.
--creds-file	INGESTR_CREDS_FILE	creds.json	Path to the JSON file used to store saved UI connections.
--logs-dir	INGESTR_LOGS_DIR	logs	Directory where per-run JSONL log files are written.
--db	INGESTR_DB	ingestr.db	Path to the SQLite database used to store run history and logs.

For example:

ingestr server \
  --port 3000 \
  --creds-file ~/.ingestr/creds.json \
  --logs-dir ~/.ingestr/logs \
  --db ~/.ingestr/ingestr.db

The same configuration can be provided with environment variables:

export INGESTR_PORT=3000
export INGESTR_CREDS_FILE=~/.ingestr/creds.json
export INGESTR_LOGS_DIR=~/.ingestr/logs
export INGESTR_DB=~/.ingestr/ingestr.db

ingestr server

Connections

The Connections tab stores named source and destination connection details. Choose a connector type, fill in the connector fields, and click Save Connection. Saved connections are written to the configured credentials file.

The credentials file is a local JSON file and is written with owner-only file permissions. Keep this file private, because it can contain passwords, tokens, or paths to credentials files.

New Runs

The New Run tab builds an ingestion run from saved source and destination connections.

The UI supports these run settings:

• Source connection
• Source table
• Destination connection
• Destination table
• Incremental strategy: replace, append, merge, or delete+insert
• Primary keys
• Incremental key

If the destination table is empty, the server uses the source table as the destination table.

For local file connectors such as CSV, SQLite, and DuckDB, absolute paths are recommended so the generated URI resolves to the intended file.

Run History

The Run History tab shows previous runs from the SQLite database configured with --db. Runs are ordered by start time, newest first. Use View Logs to expand the stored logs for a run.

Logs are stored in two places:

• The SQLite database, so they can be shown in run history.
• The logs directory as one JSONL file per run.

API

The web UI uses these local API routes:

Route	Description
GET /api/connectors	Lists connector metadata used by the UI forms.
GET /api/credentials	Lists saved connections.
POST /api/credentials	Saves a connection.
DELETE /api/credentials/{id}	Deletes a saved connection.
POST /api/run	Starts an ingestion run.
GET /api/jobs/{id}	Returns the in-memory status for an active or recently completed job.
GET /api/runs	Lists persisted run history. Supports limit and offset.
GET /api/runs/{id}	Returns a persisted run.
GET /api/runs/{id}/logs	Returns persisted logs for a run.
GET /api/ws/logs/{id}	Streams live logs for a running job over WebSocket.

Security

The server does not require authentication. Run it only on a trusted machine or trusted network, and do not expose it publicly. Saved connections can contain sensitive credentials, and anyone who can reach the UI can start ingestion jobs with those saved connections.