CLI
mink runs a node and manages the set of tables. Table commands talk to Postgres and Kafka directly, so they need no running node.
| Command | Does |
|---|---|
mink serve [-c <file>] [--roles <roles>] | Runs a node |
mink table add <topic> [-c <file>] [--granularity <g>] | Makes a topic a table |
mink table remove <topic> [-c <file>] | Removes a table. Its objects and rows are deleted after lease_ttl. |
mink table list [-c <file>] | Lists the tables and their states |
mink granularity <topic> <g> [-c <file>] | Changes a table's time partitioning for new files |
-c names a TOML config file. MINK_* environment variables override it, and --roles overrides both. See Configuration.
<g> is hour, day, month or none. Without --granularity, a table gets the granularity setting, hour by default.
Each table command has an admin API route that does the same. See Admin API.
bash
mink serve -c mink.toml --roles catalog,s3
mink table add orders -c mink.toml --granularity day
mink table list -c mink.toml