v6.4.1

[6.4.1] - 2026-04-07

Fixed

• Optimized /governance/dreps/:drep_id/delegators query. Reduces execution time and buffer hits for large dreps like drep_always_abstain

What's Changed

• chore: expose min and idleTimeoutMillis pg options by @slowbackspace in #321
• fix version attribute on nix package by @cleverca22 in #322
• fix writeShellApplication being dumb and creating a bin/blockfrost-backend-ryo-6.4.1 by @cleverca22 in #324
• chore: optimized /governance/dreps/:drep_id/delegators by @slowbackspace in #323

Full Changelog: v6.4.0...v6.4.1

v6.4.0

[6.4.0] - 2026-03-30

Changed

• Replaced idle_session_timeout (server-side Postgres setting) with idleTimeoutMillis (client-side pg-pool setting) for idle connection cleanup. Set via dbSync.idleTimeoutMs in config or BLOCKFROST_CONFIG_DBSYNC_IDLE_TIMEOUT_MS env var (value in milliseconds). Disabled by default (pg-pool falls back to its own 10,000ms default). The previous dbSync.idleSessionTimeoutMs / BLOCKFROST_CONFIG_DBSYNC_IDLE_SESSION_TIMEOUT_MS options are removed.
• Exposed min pool size option. Set via dbSync.minConnections in config or BLOCKFROST_CONFIG_DBSYNC_MIN_CONN env var. Disabled by default (pg-pool defaults to 0).

Fixed

• Fixed connection leak in /accounts/:stake_address/utxos where the DB connection was not released on 404 responses

What's Changed

• add metrics for tracking pg_pool by @cleverca22 in #320

Full Changelog: v6.3.4...v6.4.0

v6.3.4

[6.3.4] - 2026-03-17

Added

• Configurable idle_session_timeout for dbsync connections. Set via dbSync.idleSessionTimeoutMs in config or BLOCKFROST_CONFIG_DBSYNC_IDLE_SESSION_TIMEOUT_MS env var (value in milliseconds). Disabled by default

Fixed

• Fixed connection leak in /governance/proposals/:gov_action_id/metadata where the DB connection was not released on validation failure

What's Changed

• feat(health): replace [HEALTH] logs with prometheus counter health_ch… by @hlolli in #316
• chore: configurable idle_session_timeout by @slowbackspace in #317

Full Changelog: v6.3.3...v6.3.4

v6.3.3

What's Changed

• fix: timeout log even when dbsync conn succeeded by @slowbackspace in #315

Full Changelog: v6.3.2...v6.3.3

v6.3.2

[6.3.2] - 2026-03-11

Changed

• Optimized /accounts/:stake_address/addresses query by applying pagination before joining back to tx_out, significantly reducing execution time for large result sets

Fixed

• CI: Fixed NixOS integration test commands to use nix develop with flakes

What's Changed

• feat(sql): optimise stake address sql query by @hlolli in #312
• chore: release 6.3.2 by @slowbackspace in #314

Full Changelog: v6.3.1...v6.3.2

v6.3.1

[6.3.1] - 2026-03-04

Added

• Added configurable statement_timeout for DB Sync queries. Configurable via dbSync.statementTimeout in config or BLOCKFROST_CONFIG_DBSYNC_STATEMENT_TIMEOUT env var (value in milliseconds). Defaults to no timeout (PostgreSQL server default)
• Configurable slow request detection. When server.slowRequestThresholdMs (or env BLOCKFROST_CONFIG_SERVER_SLOW_REQUEST_THRESHOLD_MS) is set, a warning is logged for any request that has not completed within the given number of milliseconds. Disabled by default
• Configurable DB connectivity timeout for the /health endpoint. When server.healthCheckDbTimeoutMs (or env BLOCKFROST_CONFIG_SERVER_HEALTH_CHECK_DB_TIMEOUT_MS) is set, /health returns { is_healthy: false } instead of hanging if the DB connection takes longer than the configured value. Disabled by default
• Configurable connection timeout for acquiring a DB connection from the pool. Set via dbSync.connectionTimeoutMs in config or BLOCKFROST_CONFIG_DBSYNC_CONNECTION_TIMEOUT_MS env var (value in milliseconds). Defaults to no timeout

Changed

• Nutlink:
  • Nutlink metadata fetch errors are no longer reported to Sentry
  • Improved error message and log level changed from error to warn

What's Changed

• chore: don't report nutlink metadata fetch errors to sentry, set log … by @slowbackspace in #305
• fix: prevent handle leak and add configurable statement_timeout for D… by @slowbackspace in #308
• chore: add dbSync.connectionTimeoutMs by @slowbackspace in #309
• fix: dev build by @slowbackspace in #310
• chore: log the reason for health check fail by @slowbackspace in #311

Full Changelog: v6.3.0...v6.3.1

v6.3.0

[6.3.0] - 2026-03-02

Added

• tx_slot, block_height and block_time fields to /accounts/{stake_address}/delegations, /accounts/{stake_address}/registrations, /accounts/{stake_address}/withdrawals, /accounts/{stake_address}/mirs endpoints
• treasury_donation field to /txs/{hash} endpoint

What's Changed

• feat: add tx_slot and block_time to /accounts/:sa/*, treasury_donation in /txs/:tx_hash by @slowbackspace in #306

Full Changelog: v6.2.0...v6.3.0

v6.2.0

[6.2.0] - 2026-02-12

Added

• New registered field in /accounts/:stake_address response indicating whether the stake address is currently registered

Fixed

• pool_id in /accounts/:stake_address now correctly returns null if delegation occurred before the latest deregistration

What's Changed

• fix: active and pool_id in stake address by @slowbackspace in #301
• chore: pm2 instance limit configuration by @mkmks in #302

New Contributors

• @mkmks made their first contribution in #302

Full Changelog: v6.1.1...v6.2.0

v6.1.1

What's Changed

• chore: include pm2 instance number in databaseSyncApplicationName by @slowbackspace in #300

Full Changelog: v6.1.0...v6.1.1

v6.1.0

[6.1.0] - 2025-01-09

Changed

• Optimized /accounts/:stake_address/addresses query performance
  • Requires new index: bf_idx_tx_out_sa_paycred_script_id (see README)

What's Changed

• ci: update blockfrost-tests and lint by @hlolli in #298
• chore: faster /accounts/:stake_address/addresses by @slowbackspace in #299

New Contributors

• @hlolli made their first contribution in #298

Full Changelog: v6.0.0...v6.1.0