CrateDB n8n integrationAutomate CrateDB with n8n.
Three operations, and all the SQL you care to write. The CrateDB n8n integration gives you one node, 1 resource and a direct connection to the cluster: Execute Query runs the statement, Insert adds rows, Update rewrites the ones already stored. No CrateDB trigger exists, so the workflow starts elsewhere.
Verified Trustpilot reviews · AI, automation & growth agency
What does the CrateDB n8n integration actually do?
The CrateDB node is the piece that lets an n8n workflow talk to a CrateDB cluster in SQL. One resource, 3 operations: Execute Query runs whatever statement you write, Insert adds rows, Update rewrites rows that are already there. The credential describes a database, host and port and database name, not a web API key. That difference shapes everything else on this page.
Start with ingestion, because that is where most people meet the node. A Schedule Trigger, the n8n node that fires on a fixed interval, pulls the day's lines from Google Sheets, and an Insert operation drops them into a table. What makes or breaks that step is the Columns field: you list the property names carried by the incoming item, id,name,description, and each one lands in the column of the same name. Column types are declared in the same place, by appending :type to a name, as in id:int,name:text.
Correction is the second move. Prices change, quantities change, and yesterday's rows have to catch up. Update handles that in one pass: Update Key names the property that decides which rows get touched, normally id, while Columns names what gets rewritten. Values still come from the item, through an expression, the n8n syntax that reads a value off the incoming item, such as {{ $json.price }}.
Reading is the third, and the habit worth building lives in two fields. Put quantity,price in Query Parameters, write quantity > $1 in the Query, and the values travel beside the SQL instead of being glued into it. The result set arrives as items, ready for a Slack message or a write into another store such as MongoDB.
One point deserves to be said plainly: the CrateDB credential opens a connection to the database, not an HTTP API. The usual reflex of falling back on the HTTP Request node does not apply here. The fallback is Execute Query, which accepts any statement you can write in SQL, and that is why 3 operations are enough where a web app node needs a long list of them. If your data also lives elsewhere, there are pages for Postgres and MySQL.
Two community templates show the whole sequence. Workflow 597 creates a table in CrateDB and inserts data into it; workflow 609 handles incident reporting across PagerDuty and CrateDB. Open one next to the node and the field names stop being abstract. For the rest of n8n rather than this single node, the n8n training picks up from there.
How do you connect CrateDB to n8n?
- 01
Collect the cluster details
Open the Credentials menu in n8n and create a CrateDB credential. Three fields say where to knock: Host, the name of your instance, Port, the number it listens on, and Database, the database you want to reach. You need a running CrateDB instance first, and the n8n documentation points to CrateDB's own connection guide for the default values of these fields.
- 02
Fill in the account and the encryption
User and Password are the account n8n signs in with, so pick one that can already read and write the tables you plan to touch. SSL sets how the connection is secured and takes one of three values: Allow, Disable, Require. Choose the one your cluster is configured for instead of keeping whatever is preselected.
- 03
Select the credential in the node
Drop a CrateDB node on the canvas and pick the credential from its dropdown. A credential is created once and reused across workflows, which means the next CrateDB node already offers it without any retyping. Then choose the operation: Execute Query, Insert or Update. Running a short Execute Query before anything else is the fastest proof that the connection holds.
What can the CrateDB node do?
The CrateDB node exposes 3 operations. For each one: the node as you configure it in n8n, the required fields, and our field notes.
Execute a SQL query
executeQueryWhat you see in n8n
Notes & use cases
Whatever SQL you type runs against the connected database, and the rows come back into the workflow as items.
Key parameters
- Query: the statement to execute, required. It accepts n8n expressions and the placeholders
$1and$2, as inWHERE quantity > $1. - Query Parameters: in Additional Fields, a comma separated list of properties used as query parameters,
quantity,pricein the documented example. - Mode: also in Additional Fields. Multiple Queries is the default and sends several queries at once; Independently runs each one on its own, which works with Continue on Fail.
Insert rows in database
insertWhat you see in n8n
Notes & use cases
Every incoming item becomes a row in the table you name, and the node hands back the fields you ask for.
Key parameters
- Schema: required, the name of the schema the table belongs to.
- Table: required, the table that receives the data.
- Columns: a comma separated list of the item properties used as columns for the new rows, such as
id,name,description. Append:typeto a name to set a column's data type, as inid:int,name:text. - Return Fields: the fields the operation sends back to the workflow once the rows are written.
Update rows in database
updateWhat you see in n8n
Notes & use cases
Rows that already exist get rewritten with the values carried by the items, one matched set per run.
Key parameters
- Update Key: required, the comma separated list of properties that decides which rows in the database should be updated. Normally that would be
id. - Schema and Table: required here too, they aim the update at one table rather than the whole database.
- Columns: the properties written into the matched rows,
name,descriptionfor example. - Return Fields: what comes back afterwards, handy when the next node has to confirm the change.
Need help automating CrateDB with n8n?
A person reads every message.