cockroachdb_pool 1.1.0+1

cockroachdb_pool: ^1.1.0+1 copied to clipboard

A library to control a maximized number of connections to a CockroachDB cluster server.

Features:

• Scans nodes list (via HTTP API) periodically, automatically add/remove nodes.
• On critical errors or elevated timeouts it marks the node as bad and uses other nodes instead.

Discontinued #

package:postgres from ^3.0.0 supports connection pooling as part of the core library. Development and advanced pooling support will be migrated to that package.

Usage #

Once you've created the CrdbPool object, you can:

• Use it as PostgreSQLExecutionContext (from package:postgres).
• Use it as PgPool (from package:postgres_pool).
• Use CrdbPool.run for non-transactional batches with optional retry.
• Use CrdbPool.runTx for transactional batches with optional retry.