v2.6.1

What's Changed

• chore: bump poolboy fork by @v0idpwn in #715
• fix: handle equal syn timestamps, improve clustering tests by @v0idpwn in #719
• fix: regression when loading config by @v0idpwn in #722

New Contributors

• @nelsonmestevao made their first contribution in #718

Full Changelog: v2.6.0...v2.6.1

v2.6.0

Features

• Authentication cleartext password support - Added support for cleartext password authentication method (#707)
• Runtime-configurable connection retries - Support for runtime configuration of connection retries and infinite retries (#705)
• Enhanced health checks - Check database and eRPC capabilities during health check operations (#691)
• More consistency with postgres on auth errors - Improves errors in some client libraries (#711)

Performance Improvements

• Optimized ranch usage - Supavisor now uses a constant number of ranch instances for improved performance and resource management when hosting a large number of pools (#706)

Monitoring

• New OS memory metrics - gives a more accurate picture of memory usage (#704)
• Add a promex plugin for cluster metrics - for tracking latency and connection status (#690)
• Client connection lifetime metrics - adds a metric about how long each connection is connected for (#688)
• Process monitoring - Log when large process heaps and long message queues (#689)

Bug Fixes

• Client handler query cancellation - Fixed handling of :cancel_query when state is :idle (#692)

Migration Notes

• Instances running a small number of pools may see an increase in memory usage. This can be mitigated by changing the ranch shard or the acceptor counts.
• If using any of the new used ports, may need to change the defaults
• Review monitoring dashboards and include new metrics

v2.5.7

What's Changed

• fix: close proxy websocket upon TCP connection closing by @v0idpwn in #684

Full Changelog: v2.5.6...v2.5.7

v2.5.6

What's Changed

• chore: enable Coveralls reporting by @hauleth in #570
• chore: run tests on main branch as well by @hauleth in #571
• chore: run tests on main branch v2 (typo fix) by @hauleth in #572
• feat: use multicast instead of cast in a loop by @hauleth in #564
• docs: typo in README by @hauleth in #563
• chore: update dependencies by @hauleth in #546
• chore: ignore tests support files in coverage reports by @hauleth in #573
• test: cascade drop schema in test preparation by @hauleth in #561
• fix: more correct DB and user name handling by @hauleth in #580
• fix: lower log level when removing stopped listener by @hauleth in #577
• fix: do not delete metrics in SynHandler by @hauleth in #585
• chore: update poolboy to use supabase/poolboy by @abc3 in #587
• feat: use iodata for metrics export by @hauleth in #586
• test: allow changing verbosity in captured logs by @hauleth in #575
• test: do not fix version in migrations by @hauleth in #592
• chore: update Elixir to 1.18.2 by @hauleth in #589
• test: do not use meck when testing metrics controller by @hauleth in #593
• chore: update Erlang to 27 by @hauleth in #590
• chore: update Rustler to 0.36.1 by @hauleth in #591
• test: integration test for stats on other nodes by @hauleth in #560
• docs: typo in OpenAPI schema description by @hauleth in #596
• chore: add .tool-versions by @chasers in #600
• fix: set sync_threshold and discard_threshold by @chasers in #599
• fix: transactions connections not stopping after timeout by @hauleth in #598
• fix: do not verify TLS cert when connecting to the DB by @hauleth in #603
• fix: remaining SSL connections that need to set verify_none option by @hauleth in #605
• fix: add :verify_none to libcluster's stragy implementation by @hauleth in #607
• fix: correctly cleanup metrics tables by @hauleth in #610
• feat: add instance_id to metrics metadata by @hauleth in #613
• chore: update dependencies by @hauleth in #612
• test: use transaction connections when using transaction port by @hauleth in #597
• chore: parallel Coveralls.io reporting by @hauleth in #614
• chore: ignore tasks in coverage by @hauleth in #615
• fix: remove duplicate config entry by @hkrutzer in #618
• chore: update runner image for version_updated.yml task by @hauleth in #622
• chore: use SARIF output for Credo report by @hauleth in #621
• test: enable fail fast mode in JS tests by @hauleth in #617
• chore: add Sobelow linting by @hauleth in #620
• test: use default time limit on assert_receive in stats tests by @hauleth in #624
• chore: add mix_audit by @hauleth in #623
• test: reduce Meck usage by @hauleth in #626
• chore: cleanup appup generation script by @hauleth in #629
• chore: remove check for version updated on each PR by @hauleth in #631
• chore: remove unused module by @hauleth in #627
• feat: add location_key to global metadata by @hauleth in #632
• fix: use LOCATION_ID for cluster identification by @hauleth in #639
• chore: update release version by @hauleth in #640
• fix: use different name for host field to avoid conflicts by @hauleth in #645
• feat: improve metrics gathering by @hauleth in #647
• chore: remove commented out code by @hauleth in #648
• chore: submit Elixir dependencies to GitHub for analysis by @hauleth in #650
• chore: cleanup network metrics gathering by @hauleth in #652
• test: create separate tenant for stats tests instead of reusing existing one by @hauleth in #628
• chore: cleanup Credo issues by @hauleth in #625
• feat: use built in json module by @hauleth in #651
• feat: JSON output format for log messages by @hauleth in #653
• feat: add ipv6 support for supavisor internal database by @abc3 in #654
• chore: run security related tasks in pull_request_target event by @hauleth in #655
• chore: bump version to 2.5.2 by @hauleth in #657
• chore: update pg_query to v6.1 by @v0idpwn in #662
• docs: add environment variables documentation by @abc3 in #659
• Remove required fields from OpenAPI schema that don’t exist by @plarson in #633
• docs: fix information about prepared statement support by @TheOtherBrian1 in #468
• chore: tag flaky test as such by @v0idpwn in #664
• chore: ignore tmp dir by @v0idpwn in #665
• fix(logger): format mfas in accordance to logflare, put level in meta by @v0idpwn in #667
• chore: bump version to v2.5.3 by @v0idpwn in #668
• fix(logger): mv region, instance_id, location and nodehost outside ctx by @v0idpwn in #672
• fix(logger): logflare formatter consistency with logflare backend by @v0idpwn in #675
• fix: race on tenant startup causing subscribe failure/retry by @v0idpwn in #678
• fix: don't stop client connection when parameter status is updated by @v0idpwn in #666
• chore: update VERSION by @v0idpwn in #679

New Contributors

• @hkrutzer made their first contribution in #618
• @plarson made their first contribution in #633
• @TheOtherBrian1 made their first contribution in #468

Full Changelog: v2.0.21...v2.5.6

v2.0.21

What's Changed

• fix: set correct database on connection by @hauleth in #565
• fix: correct Dialyzer issue as db_name cannot be nil by @hauleth in #566
• fix: handle terminate message from proxy client by @abc3 in #568

Full Changelog: v2.0.18...v2.0.21

v2.0.15

What's Changed

• fix: validation prisma names by @abc3 in #547

Full Changelog: v2.0.14...v2.0.15

v1.1.70

What's Changed

• fix: start PromEx after Registries by @abc3 in #527

Full Changelog: v1.1.69...v1.1.70

v1.1.69

What's Changed

• fix: add partitions to TenantClients registry by @abc3 in #518

Full Changelog: v1.1.68...v1.1.69

v1.1.68

What's Changed

• chore: add mailmap to unify emails in log by @hauleth in #361
• docs: reformat documentation by @hauleth in #353
• Use RPC helper everywhere by @hauleth in #367
• chore: split Elixir tasks by @hauleth in #375
• fix: handle notice response during auth and terminate the creds retrieval process by @abc3 in #514

Full Changelog: v1.1.62...v1.1.68

v1.1.62

What's Changed

• fix: buffer data via cast when not in idle or busy state by @abc3 in #359

Full Changelog: v1.1.61...v1.1.62