Collect documents from an S3 bucket
Let ProcuLink collect new files from an S3 bucket.
Add a pickup
Open the flow, go to the Arrives step, and find Collect from a mailbox or a server. Select Add a pickup. In the Type list, choose S3 bucket.
Fill in the bucket
- Bucket is the bucket name.
- Prefix is optional. Type it to look only at objects whose names start with it, for example
invoices/. Leave it empty to look at the whole bucket. - Region starts as
us-east-1. Use the region of your bucket. - Custom S3 address (for R2 or MinIO) is only for services that are not Amazon S3. Leave it empty for Amazon S3.
- Access key ID and Secret access key are the key you made. The secret is stored encrypted.
Select Save.
Test the connection
Select Test connection. It lists the objects and counts the readable files that are waiting, but takes none.
Let it run
Leave On ticked. Every five minutes we list the bucket and take each object we have not seen before.
- An object is taken once. If you replace it with a new version, it is taken again.
- Objects whose names end with a slash, which are folders in the console, are ignored. So are objects of a type we do not read.
- We take up to 50 files at a time, and the rest on the next check.
- An object over 10 MB is skipped, and the pickup's last message says so.
- Objects stay in the bucket. We do not move, rename or delete them.
Good to know
Use a prefix when only part of a bucket is meant for ProcuLink. We look at everything under it.
To stop collecting, untick On, or select Remove.