8 months ago
I’m running an automation with n8n on Railway that processes XML invoices into JSON, then converts them into PDFs (DANFE), and sends them via WhatsApp Cloud API.
The main issue is that my PostgreSQL storage is filling up very quickly. I strongly suspect that n8n is storing the PDFs and JSON files inside the PostgreSQL database, which is causing the disk usage to explode over time.
Pinned Solution
8 months ago
By default, n8n stores execution data and workflow run data in the database. This includes inputs, outputs, and any data processed by nodes, so if your workflow outputs large JSON objects or binary files (like PDFs encoded in base64), they might be stored as part of the execution history in the DB.
You can either regularly clean up the old executions or give a max age to each data since from what I can see you dont intend on saving every file, it's just temporary for the conversion.
1 Replies
8 months ago
By default, n8n stores execution data and workflow run data in the database. This includes inputs, outputs, and any data processed by nodes, so if your workflow outputs large JSON objects or binary files (like PDFs encoded in base64), they might be stored as part of the execution history in the DB.
You can either regularly clean up the old executions or give a max age to each data since from what I can see you dont intend on saving every file, it's just temporary for the conversion.
Status changed to Solved chandrika • 7 months ago