Release Notes

• 1: PG Exporter v1.6.0
• 2: PG Exporter v1.5.0
• 3: PG Exporter v1.4.1
• 4: PG Exporter v1.4.0
• 5: PG Exporter v1.3.0
• 6: PG Exporter v1.2.2
• 7: PG Exporter v1.2.1
• 8: PG Exporter v1.2.0
• 9: PG Exporter v1.1.2
• 10: PG Exporter v1.1.1
• 11: PG Exporter v1.1.0
• 12: PG Exporter v1.0.3
• 13: PG Exporter v1.0.2
• 14: PG Exporter v1.0.1
• 15: PG Exporter v1.0.0
• 16: PG Exporter v0.9.0
• 17: PG Exporter v0.8.1
• 18: PG Exporter v0.8.0
• 19: PG Exporter v0.7.1
• 20: PG Exporter v0.7.0
• 21: PG Exporter v0.6.1
• 22: PG Exporter v0.6.0
• 23: PG Exporter v0.5.0
• 24: PG Exporter v0.4.1
• 25: PG Exporter v0.4.0
• 26: PG Exporter v0.3.2
• 27: PG Exporter v0.3.1
• 28: PG Exporter v0.3.0
• 29: PG Exporter v0.2.0
• 30: PG Exporter v0.1.2
• 31: PG Exporter v0.1.1
• 32: PG Exporter v0.1.0
• 33: PG Exporter v0.0.5
• 34: PG Exporter v0.0.4
• 35: PG Exporter v0.0.3
• 36: PG Exporter v0.0.2
• 37: PG Exporter v0.0.1

1 - PG Exporter v1.6.0

PG Exporter v1.6.0 separates health probes from collection, bounds overlapping metrics requests, preserves completed bounded refreshes after client cancellation, and corrects cross-database progress labels. It also supplies container trust roots and updates dependencies. The Go toolchain remains 1.27.3 and the PostgreSQL driver remains lib/pq 1.12.3.

Changes

• Independent health probes: slow metrics collection no longer blocks the background probe. Completing a scrape no longer overwrites a newer reachability or role observation. After initialization, HTTP health is written only by the probe; pg_up and pg_in_recovery remain collection results. A reachable database can return /up HTTP 200 even when a fatal collector makes pg_up zero.
• Controlled overlapping scrapes: at most two metrics requests are admitted, including waiting and response writing. Collection remains serial. The second request can wait and reuse caches; extra requests receive HTTP 503. Waiting can be canceled without starting SQL. The request budget is 30 seconds, shortened by a valid positive finite Prometheus scrape-timeout header.
• Collector-level cancellation: request cancellation prevents the next collector or database from starting. A collector already running with a positive timeout continues its predicates and main SQL under one shared deadline, then caches its result normally. This prevents a departing client from turning a healthy bounded refresh into a failure cached for other scrapers. A negative timeout removes the collector deadline and lets the request cancel its SQL.
• Clearer interruption accounting: interrupted passes have an INFO interrupted summary and do not increment exporter/server scrape-error counts. Availability preserves the previous value, or becomes true after a successful precheck. Non-interrupted connection and fatal collector failures still count. Query-error statistics and fatal-query ERROR logs remain unchanged.
• Correct progress labels: pg_vacuuming, pg_indexing, and pg_clustering retain cluster-wide coverage and three queries per refresh. Relations in the connected database and system catalogs resolve to names; user relations in other databases use numeric OIDs in relname, retaining datname and pid. An unrelated local relation can no longer supply a misleading name. Discovery filters or a lack of CONNECT do not filter out visible progress rows.
• Container trust: both scratch-image build paths include the system CA bundle and license. Public CA verification can use the bundled trust store. Private CAs still require an explicit mount and sslrootcert; the default TLS mode is unchanged.
• Dependencies and validation: YAML moves from v3 to v4 with the existing collector syntax retained. Prometheus common/procfs, golang.org/x/net, and supporting dependencies are updated. CI verifies container payload/TLS behavior and inspects all DEB architectures, in addition to RPM compatibility and checksums.
• Collector descriptions: help text now distinguishes I/O bytes and operations, completed checkpoints, vacuum dead-tuple memory, and CLUSTER progress. SQL metric names, label keys, types, and units are unchanged on the same PostgreSQL version.

Upgrade notes

1. Upgrade the configuration with the binary. Merge the new pg_exporter.yml or split YAML files; package upgrades preserve local configuration. A binary-only upgrade does not fix progress SQL. The bundle still contains 58 YAML files, 96 branches, 59 namespaces, and 460 distinct metric definitions, including disabled collectors and two unexpanded histograms.
2. Check progress-label consumers. Cross-database relname values can change from an incorrect name to an OID, creating new time series. Preserve database and task identity in dashboards. Names in the connected database and names of system catalogs remain readable.
3. Choose the right health signal. /up and role endpoints reflect the latest health probe. pg_up reflects main-target collection, including fatal collector failures. Non-fatal and child-database errors require their query/server metrics. --disable-intro hides all built-in metrics, including pg_up, while HTTP health remains available.
4. Allow the PgBouncer probe connection. The main PgBouncer pool maximum changes from one to two connections. PostgreSQL keeps three on the main target and one per discovered database.
5. Keep positive timeouts on fatal collectors. Canceling a negative-timeout fatal query caches its failure for the normal TTL. A subsequent live scrape can count that cached failure and return pg_up 0 until a successful refresh. TTL zero or --disable-cache still forces a refresh on every scrape.
6. Review scrape concurrency and budgets. Excess requests or interrupted collection can return 503. A started bounded collector can extend work beyond the request budget, normally by up to its own timeout. Network stalls can last longer; neither timeout is a hard wall-clock guarantee, and connection handshakes need an appropriate DSN connect_timeout.

A real error coinciding with request cancellation can be omitted from the two pass-error counters because classification uses both the pass error and the request context. Query-level failures remain visible. A first scrape interrupted before a successful precheck leaves pg_up at zero until later recovery. Failure TTL, reload behavior, and immediate SIGTERM exit remain unchanged; this release does not add background refresh or graceful shutdown.

For upgrades from v1.4.1 or earlier, the v1.5.0 migration notes still apply, including subscription labels, receiver-state mappings, and PostgreSQL 10–13 monitoring grants. See metric compatibility, HTTP API, and scrape lifetime for the operating contracts.

Verification scope

Release validation covers the full race suite, both HTTP mux modes, module/generated-config consistency, static checks, six target builds, container TLS/payload checks, and package validation. Three cancellation regressions also passed 200 consecutive race-detector runs. The earlier reverse control failed with the interruption fix removed. On PostgreSQL 18.6, leaving during a bounded 0.8-second fatal query kept both pass-error counters at zero; the next scrape returned pg_up 1 and the cached metric without repeating the SQL.

Local acceptance of the exact release artifacts also covered PostgreSQL 9.2–9.6 with the legacy configuration, PostgreSQL 10–18 and 19 RC1 with the default configuration, and PgBouncer 1.26.1. Linux amd64/arm64/ppc64le archives and RPM/DEB packages passed runtime and installation checks; ARM64 package upgrades from 1.5.0 preserved local configuration. Darwin ARM64 passed 24 functional groups, including HTTPS authentication, reload overlap, standby promotion, and reconnect. Windows and Intel macOS received static checks only; container-based package tests did not exercise a real systemd boot. The published Docker Hub 1.6.0 and latest tags resolve to the same amd64/arm64 manifest. Both registry images were pulled and passed CA, payload, and TLS checks; their binaries match the corresponding release archives byte for byte.

Progress SQL variants were checked as a monitoring role on PostgreSQL 12, 17, and 18; the system-catalog follow-up and real cross-database VACUUM were verified on 18.6. These are scoped acceptance results, not a new exhaustive audit of every PostgreSQL release, extension, dashboard, or production workload. Composite, Patroni, pgBackRest, and PostgreSQL-log components remain on the separate snort development line and are not included in v1.6.0.

Checksums

SHA256 checksums

2 - PG Exporter v1.5.0

PG Exporter v1.5.0 updates the Go toolchain and dependencies, fixes replication collector identity and state mappings, and closes six runtime defects around target URLs, discovery, health, collector planning, durations, and HTTP routing.

The bundle still contains 58 YAML files, 96 collector branches, 59 namespaces, and 460 distinct metric definitions. The metric count spans all version branches, including disabled collectors and two unexpanded SQL histograms; it is not a promise of 460 series in each scrape.

Changes

• Safer connection URLs: malformed URL syntax, query escaping, or unsafe decoded parameter names stop startup without exposing the original connection string. Explicit TLS options can no longer disappear silently when malformed parameters are normalized. An empty or whitespace-only selected PG_EXPORTER_URL_FILE also fails startup.
• Correct database discovery: database identity follows lib/pq, including the dbname and database aliases. Discovery removes conflicting database parameters when substituting a database path and preserves special characters. This prevents repeated collection of one database, duplicate samples, and resulting HTTP 500 responses.
• Consistent health responses: each response derives its status code and body from one atomic state snapshot. Concurrent probes, scrapes, and reloads can no longer combine HTTP 200 with a down or starting body. Health still reports cached target state.
• Collector plans follow metadata: changes to the connected user, schema set, or extension set invalidate the plan after a successful metadata precheck. New eligible collectors appear on the next scrape; removed dependencies stop selecting collectors. Unchanged sets preserve their caches.
• Bounded configuration durations: YAML TTLs, timeouts, and predicate TTLs reject NaN, infinity, and positive duration overflow. Zero/default semantics and finite negative timeouts that disable deadlines remain unchanged.
• Reachable metrics paths: startup checks the actual metrics handler selected by the active HTTP mux, including decoded paths and redirects. A path cannot pass validation while routing to the homepage or a redirect.
• Replication identity and states: every pg_sub_* sample gains a datname label, distinguishing subscriptions with the same name in different databases. pg_recv_state uses the actual receiver states, and its legacy query uses PG9.6-compatible substring().
• More precise collector descriptions: document optional replication-origin permissions, publication table counts versus subscription table/sequence counts, and pg_backend grouping by backend type. These description changes do not add metric families or enable optional collectors.

Metric and dashboard compatibility

Compared with v1.4.1 on the same PostgreSQL version, v1.5.0 does not add, remove, or rename bundled SQL metric names. Types and units remain unchanged, but labels and state mappings do change. PostgreSQL 19 support first shipped in v1.3.0; upgrading the database can also change the selected SQL branch and a metric’s meaning.

Users upgrading from PG17 or earlier must also migrate older names: checkpoint metrics left pg_bgwriter_* in PG17, and pg_wal_write, pg_wal_sync, pg_wal_write_time, and pg_wal_sync_time are absent from PG18 onward. The full compatibility guide lists each replacement, label/aggregation differences, and mixed-version PromQL.

Before rollout, evaluate the actual dashboard and alert expressions and test missing-series input. A successful scrape does not ensure every expected metric exists; an empty alert expression can silently stop firing. The missing-metric checks include a per-target rule and explain when absence is expected. The full PG19 metric table links to the tagged collector definitions; PostgreSQL validation remains scoped to 19 RC1.

Toolchain and dependencies

The database driver remains lib/pq 1.12.3. There is no driver migration or new optional data source in this release. Prometheus client/model/procfs, YAML, compression, OAuth2, and other transitive modules are also refreshed; a version bump alone is not evidence that an application CVE was fixed.

Security updates

The upgrade incorporates upstream security fixes; it does not mean every listed CVE was exploitable through PG Exporter. The comparison below uses the v1.4.1 source and dependencies with Go 1.26.5. S means a static symbol match, P an imported-package match without an affected-symbol call, and M a module match whose affected package is not imported. Symbol matches can be conservative and are not attack reproductions.

Go standard library and build tools

Go 1.27.3 includes CVE-2026-111943, which adds an X.509 denylist for untrusted intermediate certificates, plus cumulative HTTP, TLS, template, and parser fixes. The following affected standard-library entries were present in the old baseline; the match column is from Darwin/arm64 and platform-specific preconditions still apply.

The build toolchain also includes module/toolchain download-validation fixes CVE-2026-56865, CVE-2026-56864, CVE-2026-94444, and CVE-2026-94447. These strengthen the build environment, rather than describing exporter runtime vulnerabilities. The official announcements cover Go 1.26.6, Go 1.27.2, and Go 1.27.3.

Go modules

On 2026-10-23, govulncheck found no symbol-level or imported-package vulnerabilities in the v1.5.0 candidate for Darwin/arm64, Linux/amd64, and Windows/amd64. The unimported OpenPGP package still produces the module-only advisory GO-2026-5932, which has no fixed version. This is not a claim that every dependency module is free of advisories. Upgrading Go alone while retaining the old modules leaves the five x/net matches; both the compiler and dependency changes matter.

Use the rebuilt release artifacts: installing a newer Go compiler on a host does not change an already compiled exporter binary.

Upgrade notes

1. Deploy the matching collector configuration. Upgrading only the binary does not apply SQL or label corrections. Compare local customizations before replacing /etc/pg_exporter.yml.

2. Review the pg_sub_* label change. The new datname label creates new time series. Keep database identity in PromQL joins and aggregations where needed. pg_sub_id remains a Gauge.

3. Grant the PG10–13 subscription prerequisite. As an administrator, run the following in each database used as the exporter’s main connection database, using the actual monitoring role. Granting it only in a subscription’s database is insufficient. It does not expose subconninfo; PG14+ needs no extra grant under standard catalog ACLs.

   GRANT SELECT (oid) ON pg_catalog.pg_subscription TO monitor;

4. Check receiver-state alerts. streaming=0, stopping=4, connecting=5 (PG19+), starting=6, waiting=7, restarting=8, stopped=9, and unknown/NULL =-1; codes 1–3 are reserved. A missing receiver row remains absent.

5. Correct previously tolerated invalid input. Broken URL escapes, empty selected URL files, non-finite durations, and unreachable metrics paths now fail validation. Prefer /metrics for compatibility with both HTTP mux implementations.

Omitted sslmode still defaults to disable; configure TLS explicitly for remote connections. The scratch container still requires a mounted CA for certificate verification. Empty YAML reloads still apply zero configured collectors, invalid files in a directory may be skipped, and SIGTERM does not wait for active HTTP requests. See Security, HTTP API, and Bundled Collectors for the operating contracts.

Validation scope

Pre-release validation covered the race suite, both HTTP mux modes, static analysis, module and generated-config consistency, roughly 604,000 short-run fuzz cases, and six target builds. Non-superuser default collection passed on PostgreSQL 9.1–9.6, 10–18, 19 RC1, and PgBouncer 1.26.1. Runtime checks included conflicting database aliases, special database names, schema/extension changes, failed reloads, reconnect recovery, physical standby promotion, and concurrent HTTP requests. RPM/DEB installation and Docker builds were also checked.

These checks cover the recorded versions and scenarios, not every extension combination or production workload. Composite, Patroni, pgBackRest, and PostgreSQL-log components remain development work outside v1.5.0.

Checksums

SHA256 checksums

3 - PG Exporter v1.4.1

v1.4.1 is a maintenance release focused on metric accuracy and RPM upgrade compatibility.

Highlights

• Fix pg_subrel_count overcounting when parallel logical replication apply workers are active: deduplicate pg_stat_subscription by subscription ID and name before aggregating subscription relation states
• If a query result omits a configured LABEL column, that collector’s scrape now fails atomically instead of emitting an empty label or retaining stale results; other non-fatal collectors continue normally
• Rename the official RPM package and artifact prefix from pg_exporter to pg-exporter; the new package provides and obsoletes the legacy name, allowing direct upgrades
• Refresh version and standalone package metadata, add missing-label regression coverage, and validate RPM configuration, artifact names, and compatibility metadata in CI

Upgrade Notes

• Custom collector SQL must return every LABEL column; the result schema must remain complete even when the query returns zero rows
• If automation matches GitHub RPM file names directly, update pg_exporter-*.rpm to pg-exporter-*.rpm

Checksums

SHA256 checksums

4 - PG Exporter v1.4.0

v1.4.0 introduces the Snapshot Histogram metric type and the new pg_xact_age transaction age collector, along with a round of systematic hardening across HTTP routing, packaging, and the build toolchain.

New Features

• New HISTOGRAM column type: SQL observations are grouped by labels into snapshot distributions. The _bucket, _count, and _sum families are Gauges. Bucket bounds must be finite and strictly increasing; +Inf is added automatically. The le label is reserved, and reload replaces the bucket layout
• New pg_xact_age collector: exposes the distribution of open transaction age (pg_xact_age_seconds) and idle-in-transaction age (pg_xact_age_idle_seconds) as histograms; cluster-level collection, client backends only, 10s TTL
• Default config bundle grows from 57 to 58 definition files; pg_xact_age takes slot 0450, with pg_lock / pg_lock_stat / pg_query renumbered to 0460 / 0470 / 0480 (contents unchanged)

Fixes & Improvements

• HTTP route registration is isolated to a private ServeMux instead of the global DefaultServeMux, so endpoints registered by third-party libraries can no longer be exposed accidentally
• --web.telemetry-path is strictly validated at startup: empty paths, paths not starting with /, paths containing ? # { }, conflicts with built-in endpoints, and non-canonical paths like //metrics that could never be matched all fail fast with a clear error
• The landing page HTML-escapes the telemetry path
• The primary connection pool is properly closed when --fail-fast pre-check fails
• RPM / DEB packaging fixes (#105): the prometheus system user’s HOME now points to /var/lib/prometheus (where libpq looks for ~/.pgpass), and the packaged default URL gains the /postgres database name so libpq no longer falls back to a database named after the OS user
• Version string unified with the v prefix: --version output, the /version endpoint, and the pg_exporter_build_info{version=...} label of official release binaries now report v1.4.0 (GoReleaser artifacts previously reported unprefixed 1.3.0); artifact file names, package versions, and Docker image tag conventions are unchanged
• Histogram value casting aligned with the scalar path: timestamp and boolean columns are exempt from scale
• GoReleaser embedded package metadata corrected: supported range updated to PostgreSQL 9.x - 19+ and PgBouncer 1.8 - 1.25+
• make docker restores GOPROXY / GOSUMDB build-arg passthrough

Engineering & Build

• Build toolchain updated to Go 1.26.5, exporter-toolkit v0.17.1, and prometheus/common v0.70.0
• New regular CI verification workflow: module tidy check, generated-config drift check (make conf output must stay in sync with config/*.yml), race tests, and six-platform cross-builds
• Docker build tooling consolidated: the docker/ scripts and make docker-release are removed; multi-arch release images are built by GoReleaser
• Config coverage tests generalized: tests no longer assume specific collectors exist in the config directory, making trimmed custom bundles easier to maintain

Upgrade Notes

• If you were running with a non-canonical telemetry path (e.g. //metrics), the process will now refuse to start — previously it started but the metrics endpoint was silently unreachable

• Automation that parses the version label of pg_exporter_build_info or the --version output needs to accommodate the v prefix

• When any HISTOGRAM collector is configured, le can no longer be used as a constant label

• Query snapshot buckets directly with histogram_quantile(); do not apply rate() or increase(). See snapshot Histogram semantics.

Checksums

SHA256 checksums

5 - PG Exporter v1.3.0

v1.3.0 adds PostgreSQL 19 support, and refreshes the default collector bundle, build toolchain, and config coverage tests.

Highlights

• Default support range extended from PostgreSQL 10-18+ to 10-19+
• Default config bundle includes 57 config/*.yml definition files
• Build dependencies updated to Go 1.26.4, lib/pq v1.12.3, Prometheus client v1.23.2, and exporter-toolkit v0.16.0
• pg_recovery_state: collect pg_stat_recovery on recovery nodes, exposing promotion trigger status, replay LSN, timeline, recovery transaction time, and pause state
• pg_lock_stat: collect PG19 pg_stat_lock, exposing waits, wait time, and fast-path overflow counts by locktype
• pg_vacuum_score: collect the current-database summary from pg_stat_autovacuum_scores, exposing max autovacuum score and candidate table count
• pg_wal_19: expose wal_fpi_bytes on PG19 as pg_wal_fpi_bytes
• pg_sub_19: adapt sequence sync and logical replication conflict stats from pg_stat_subscription_stats, while keeping sync_error_count for legacy dashboards
• pg_recv: recognize the PG13+ WAL receiver connecting state on PG19
• pg_slot: recognize the PG19 replication slot invalidation reason idle_timeout
• pg_db_confl: switch to an explicit column list to avoid accidentally exporting future view columns
• pg_backup, pg_vacuuming, and pg_clustering continue reusing stable existing branches instead of changing the metric surface for low-value PG19 fields

Checksums

SHA256 checksums

6 - PG Exporter v1.2.2

Changes

v1.2.2 updates release builds to Go 1.26.2. Exporter features and configuration are unchanged.

Checksums

SHA256 checksums

7 - PG Exporter v1.2.1

v1.2.1 updates the build toolchain, standardizes configuration formatting, and adds configuration consistency tests. Collector behavior is unchanged.

Highlights

• Refresh the build toolchain: bump both release workflows and Docker build images to Go 1.26.1
• Standardize config style: switch inline description values in both current and legacy configs to double-quoted form, and regenerate merged pg_exporter.yml / legacy/pg_exporter.yml
• Add config consistency tests: verify split and merged configs remain equivalent, and check inline metric description style to reduce configuration drift
• Refresh packaging metadata: update RPM / DEB support descriptions to PostgreSQL 9.x - 18+ and PgBouncer 1.8 - 1.25+, and refresh Pigsty documentation links

Checksums

8 - PG Exporter v1.2.0

v1.2.0 updates startup, hot reload, health checks, and configuration validation. It also adds a configuration bundle for PostgreSQL 9.1–9.6.

New Features

• Add robust hot reload workflow: support platform-specific reload signals (SIGHUP / SIGUSR1) and strengthen POST /reload to refresh configs and query plans without process restart
• Switch startup to non-blocking mode: HTTP endpoints come up first even when target precheck fails, making recovery and monitoring integration smoother
• Add PostgreSQL 9.1-9.6 legacy config bundle: provide legacy/ configs and a make conf9 target for easier onboarding of EOL PostgreSQL versions
• Rework health probing architecture: use cached health snapshots with periodic probes for more consistent role-based health endpoints and smoother reload behavior
• Improve release engineering baseline: run go test and go vet in release workflows and bump build toolchain to Go 1.26.0

Bug Fixes

• Fix multiple config parsing edge cases: reject malformed metrics entries, return explicit errors when config dirs fail to load valid YAML, and harden runtime fallbacks
• Fix CLI bool flag parsing to correctly handle --flag=false style arguments
• Fix /explain output/rendering behavior by adjusting content type handling and using safer template rendering
• Fix predicate query and PG URL handling details: better BOOL/BOOLEAN predicate support, safer row lifecycle handling, and improved dbname query-parameter parsing/redaction
• Fix resource cleanup when auto-discovered targets are removed by closing dropped server connections asynchronously
• Fix metric/label validation details including const-label conflict checks, scaled default-value handling, and Prometheus naming/rule checks

Checksums

9 - PG Exporter v1.1.2

Changes

Fixed the pg_timeline configuration and updated Go dependencies.

Checksums

SHA256 checksums

10 - PG Exporter v1.1.1

Minor release with new collectors and bug fixes.

New Features

• New pg_timeline collector for timeline monitoring
• New pg_sub_16 collector branch to exclude parallel operations in subscriptions (PostgreSQL 16+ compatibility)

Bug Fixes

• Fix: Add coalesce for slotname in pg_recv collector to handle NULL values

Checksums

SHA256 checksums

11 - PG Exporter v1.1.0

Updated dependencies and built with Go 1.25.5.

Collector Changes

• pg_setting: Major refactor for PG10-18 compatibility with missing_ok support
  • Add 15 new metrics: max_parallel_workers, max_parallel_workers_per_gather, max_parallel_maintenance_workers, shared_buffers, maintenance_work_mem, effective_cache_size, fsync, full_page_writes, autovacuum, autovacuum_max_workers, checkpoint_timeout, checkpoint_completion_target, hot_standby, synchronous_commit, io_method
  • Rename work_memory_size to work_mem
  • Change min_version from 9.6 to 10, explicit ::int type casting
• pg_size: Fix log directory size detection, use logging_collector check instead of path pattern matching
• pg_table: Performance optimization, replace LATERAL subqueries with JOIN for better query performance; fix tuples and frozenxid metric type from COUNTER to GAUGE; increase timeout from 1s to 2s
• pg_vacuuming: Add PG17 collector branch with new metrics indexes_total, indexes_processed, dead_tuple_bytes for index vacuum progress tracking
• pg_query: Increase timeout from 1s to 2s for high-load scenarios
• pg_io: Fix typo in reuses description (“in reused” -> “is reused”)
• pg_checkpointer: Fix description for pg_checkpointer_10 (“9.4+” -> “9.4-17”)
• pg_db_confl: Fix description for pg_db_confl_15 (“9.1 - 16” -> “9.1 - 15”)
• Format alignment fixes for pg_db, pg_indexing, pg_clustering, pg_backup

Other Changes

• Fix release year by @anayrat

Checksums

SHA256 checksums

12 - PG Exporter v1.0.3

Changes

• Updated dependencies and built with Go 1.25.4.
• Fix #80 Conflict with libpq env variables
• Change default value of auto-discovery to true by @kadaffy

Checksums

SHA256 checksums

13 - PG Exporter v1.0.2

Changes

• Updated dependencies and built with Go 1.25.0.
• Added the PG Exporter website.
• Release with GoReleaser for more operating systems and architectures with CI/CD pipeline:
  • add windows amd64 support
  • add linux ppc64le support

Checksums

SHA256 checksums

14 - PG Exporter v1.0.1

Changes

• Add Docker Hub images: pgsty/pg_exporter
• Updated Go dependencies and built with Go 1.24.5.
• Disable pg_tsdb_hypertable collector by default, since timescaledb catalog is changed.

Checksums

15 - PG Exporter v1.0.0

Changes

Added PostgreSQL 18 collector branches:

Checksums

MD5 checksums published in the GitHub Release:

16 - PG Exporter v0.9.0

Default Collectors

• new metrics collector for timescaledb hypertable
• new metrics collector for citus dist node
• new metrics collector for pg_wait_sampling wait event profile
• pg_slot overhaul: Add 16/17 pg_replication_slot metrics
• allow pg_slot collector run on replica since 16/17
• refactor pg_wait collector to agg from all processes
• restrict pg_clustering, pg_indexing, pg_vacuuming run on primary
• mark all reset_time as GAUGE rather than COUNTER
• fix pg_recovery_prefetch_skip_fpw type from GAUGE to COUNTER
• fix pg_recv.state type from LABEL to GAUGE
• Format collector in compact mode
• new default metric pg_exporter_build_info / pgbouncer_exporter_build_info
• add server_encoding to pg_meta collector

• Added 12 pg_setting metrics: wal_block_size, segment_size, wal_segment_size, wal_level, work_memory_size, shared_memory_size, hugepage_count, hugepage_status, max_wal_size, min_wal_size, max_slot_wal_keep_size, archive_mode.

Exporter Codebase

• normalize collector branch name with min pg ver suffix
• Add license file to binary packages
• Moved the repository from Vonng/pg_exporter to pgsty/pg_exporter.
• refactor server.go to reduce Compatible and PostgresPrecheck complexity
• rename metrics collector with extra number prefix for better sorting
• Updated dependencies.
• execute fatal collectors ahead of all non-fatal collectors, and fail fast

17 - PG Exporter v0.8.1

Changes

• Updated dependencies.
• Bump golang.org/x/net from 0.35.0 to 0.36.0 #67
• Update docker images building tags

18 - PG Exporter v0.8.0

Changes

• Add PgBouncer 1.24 new metrics support (stat, pool, database)
• Fix: 310-pg_size.yml fails if log dir not set properly #64 by @Süleyman Vurucu
• Build with the latest Go 1.24 and bump all dependencies
• Refactor logging with the standard log/slog instead of go-kit Full changelog

19 - PG Exporter v0.7.1

Changes

Routine update with dependabot

• Feat: support specifying configuration as Reader by @ringerc in #62
• Bump golang.org/x/crypto from 0.21.0 to 0.31.0 by @dependabot in #63
• Fix some typos Full changelog

20 - PG Exporter v0.7.0

Changes

Updated the codebase for Go 1.22.4.

• PostgreSQL 17 Metrics Support by @Vonng

• pg_exporter: predicate queries feature by @ringerc

• Do a clean build in the dockerfile by @ringerc

• pg_exporter: don’t panic after “bind: address already in use” by @ringerc

• pg_exporter: fix /stat endpoint formatting by @ringerc

• pg_exporter: omit default query properties on yaml export by @ringerc

• Exclude template DBs from discovery and schema-qualify discovery query by @ringerc

• Fix some typos and some metric description mistakes by @ringerc

• The proposed pgx driver migration (#52, @ringerc) was reverted before release. This tag uses lib/pq.

• Added PgBouncer 1.23 metrics (source commit).

• Corrected log caller attribution (#55, @ulysseskk).

21 - PG Exporter v0.6.1

Changes

• Updated golang.org/x/net from v0.10.0 to v0.17.0 and golang.org/x/crypto to v0.17.0 for security fixes (#38 covers x/crypto)
• Fix connect-timeout propagation (#37), contributed by @mouchar

Checksums

The original release published MD5 checksums for its archives and packages:

22 - PG Exporter v0.6.0

Changes

• Updated dependencies to address reported vulnerabilities.
• Added PostgreSQL 16 collectors and PostgreSQL 15 compatibility fixes (#35, @japinli).
• Added ARM64 packages.
• Removed the fixed monitor schema from pg_query queries. pg_stat_statements must be visible through search_path, such as when installed in public.
• Accepted .yml as well as .yaml in configuration directories (#34, @japinli).
• Changed PgBouncer version-parsing messages from info to debug level.
• Added the missing relid column to pg_table_10_12.

23 - PG Exporter v0.5.0

Exporter changes

• Used nFPM to build RPM and DEB packages.
• Added column.default to substitute a value for NULL.
• Added column.scale to scale numeric values, such as microseconds to seconds.
• Fixed /stat output.
• Added the pgsty/pg_exporter Docker image.

Collector changes

• Converted time metrics to seconds in pg_wal, pg_query, pg_db, pg_bgwriter, pg_ssl, and pgbouncer_stat.
• Moved pg_class metrics into pg_table and pg_index.
• Enabled pg_table_size by default.

24 - PG Exporter v0.4.1

Changes

• Excluded Citus and TimescaleDB internal schemas from object monitoring.
• Removed duplicate pg_statio rows.
• Added PgBouncer 1.16 support.
• Fixed overlapping pg_repl collector branches on PostgreSQL 12.
• Added -T, --connect-timeout, and PG_EXPORTER_CONNECT_TIMEOUT to control connection timeouts, including for remote targets.
• Renamed the RPM configuration file from pg_exporter.yaml to pg_exporter.yml.

25 - PG Exporter v0.4.0

Changes

• Added PostgreSQL 14 support.
• Revised the default metric configuration while preserving support for older configuration files.
• Added --auto-discovery, --include-database, and --exclude-database.
• Added multi-database collection through automatic discovery.

26 - PG Exporter v0.3.2

Changes

• Fixed an edge case in redacted connection strings.
• Corrected documentation and spelling errors.

27 - PG Exporter v0.3.1

Changes

Fixed default-configuration errors, especially on PostgreSQL versions below 13:

• Restricted primary_conninfo collection to the supported collector branch.
• Added funcid to pg_func labels to distinguish functions with the same name.
• Corrected the program name in version output to pg_exporter.

28 - PG Exporter v0.3.0

Changes

• Updated the default configuration for PostgreSQL 13 metrics, including pg_slru, pg_shmem, pg_query13, and pg_backup.
• Added HTTP endpoints for health and recovery-state checks.
• Added a startup HTTP server that reports pg_up 0 until initialization finishes.
• Added sslmode=disable when the connection URL omits sslmode.
• Fixed bugs and spelling errors.

29 - PG Exporter v0.2.0

Changes

• Added RPM packages for YUM and a Linux service definition.
• Added the skip option to query configuration.
• Fixed pgbouncer_up.
• Added configuration reload support.

30 - PG Exporter v0.1.2

Changes

• Fixed pgbouncer_up.
• Added dynamic configuration reload.
• Removed shard-related logic.
• Added a bulky mode to the default configuration.

31 - PG Exporter v0.1.1

Changes

• Fixed a startup hang when a query failed.

32 - PG Exporter v0.1.0

Changes

The first stable release.

33 - PG Exporter v0.0.5

This is an early repository tag without a separate GitHub Release object. This archival note is based on the tag commit and historical release summary.

Changes

• Add per-collector query timeouts
• Replace the earlier skip_error behavior with the fatal collector policy
• Revise collector priority semantics
• Fix the built-in pg_up metric
• Add documentation for each bundled collector

34 - PG Exporter v0.0.4

Changes

Tested across more than 200 production nodes for about two weeks.

35 - PG Exporter v0.0.3

Changes

Passed production testing while the project remained under active development.

36 - PG Exporter v0.0.2

Changes

An early version made available for trial use.

37 - PG Exporter v0.0.1

This early version has a Git tag but no separate GitHub Release. The note is based on the tag and historical release summary.

Changes

• Added PgBouncer mode.