3 Commits

Include renames

Author	SHA1	Message	Date
lukaszraczylo	0e8a8fe33f	fix(websocket): tear down peer connection and bound backend messages ... Two defects in the bidirectional proxy: 1. Goroutine/socket leak on client disconnect. proxyClientToBackend and proxyBackendToClient each block on their own Read; when the client drops, the client-direction goroutine returns but the backend-direction goroutine stays blocked on backend.ReadMessage, so wg.Wait() never returns and the goroutine plus backend socket leak until the backend independently closes or sends. Each direction now closes the opposite connection on exit, unblocking the peer. 2. Backend->client messages were unbounded. client.SetReadLimit only bound what the client could send; a large or misbehaving backend could push arbitrarily large frames through. backend.SetReadLimit now bounds it. Tests: TestWebSocketProxy_TeardownOnClientDisconnect (backend must be closed after client disconnect), TestWebSocketProxy_BackendReadLimit (connection torn down when backend exceeds the limit).	2026-08-13 03:10:04 +01:00
lukaszraczylo	55fc2ae1de	feat(deps)!: migrate from fiber v2 to fiber v3 ... BREAKING CHANGE: upgrades the HTTP framework from gofiber/fiber/v2 to gofiber/fiber/v3, and gofiber/websocket/v2 to gofiber/contrib/v3/websocket. - Handlers now take fiber.Ctx (interface) instead of *fiber.Ctx. - DisableStartupMessage moved from fiber.Config to fiber.ListenConfig on the proxy/api/monitoring Listen calls. - cors AllowOrigins is now []string; c.BodyParser -> c.Bind().Body(). - app.Test takes fiber.TestConfig instead of an int timeout (tests updated). - WebSocket swapped to contrib/v3/websocket (same Conn/New/Config/IsCloseError/IsWebSocketUpgrade API; Conn.Query preserved). - Adopts fasthttp v1.71.0 (pulled by fiber v3), resolving the v1.71 Host-header enforcement that broke the prior pinned-v1.69 workaround. fix(connpool): the background keep-alive goroutine no longer reads the mutable global cfg (captures HostGraphQL/HealthcheckGraphQL at pool creation). Under v3 timing it raced with parseConfig/test cfg reassignment; go test -race ./... is now clean (0 data races). test: make the circuit-breaker cache-fallback and request-coalescing integration tests deterministic (cache TTL outlives the retry loop; coalescing asserts against a measured single-request baseline instead of a hard-coded count); drop the flaky timing-ratio assertion in TestTimingAttackResistance (constant-time guarantee comes from subtle.ConstantTimeCompare, not wall-clock timing). Verified: go build, go vet, golangci-lint (0 issues), go test -race ./... (all packages pass, 0 races), govulncheck (no vulnerabilities). App + dashboard + WebSocket verified live under v3 in single-node and cluster modes.	2026-06-21 13:15:14 +01:00
lukaszraczylo	c2c75d69c0	perf+coverage: optimisation pass + coverage push to ≥70% ... Performance / resource usage: - circuit_breaker_metrics: fix data race on failCounters map (RWMutex + double-checked locking) - server.go: drop user_id and op_name metric labels (Prometheus cardinality bound); de-duplicate extractUserInfo - graphql.go: gate runtime.ReadMemStats per-request behind ENABLE_ALLOCATION_TRACKING flag (default off) - graphql.go: collapse two-pass AST scan into single pass; lower-case once - sanitization.go: cache compiled redaction regexes per pattern via sync.Map; hoist inner constants to pkg vars - proxy.go: hoist connection/timeout substrings to pkg vars; sentinel errors for static error paths; drop dead Headers map alloc - metrics_aggregator.go: log-field allocation guarded by Logger.IsLevelEnabled - logging/logger.go: add IsLevelEnabled helper - lru_cache.go: 16-shard sharding, FNV-1a routing (concurrent throughput +22%) - cache/memory/lru_memory_cache.go: gzip compress/decompress moved outside mu.Lock - rps_tracker.go: RWMutex+uint64 -> atomic.Uint64 - retry_budget.go: drop unused mutex - api.go: bannedUsersIDs map+RWMutex -> sync.Map (+ snapshot/replace helpers) - tracing/tracing.go: pkg-level constSpanAttrs, copy-then-append in StartSpanWithAttributes - admin_dashboard.go: handleStatsWebSocket reuses bytes.Buffer + json.Encoder per connection Build / runtime: - Makefile: -ldflags="-s -w" -trimpath, CGO_ENABLED=0 for build (=1 for test recipes) - Dockerfile + Dockerfile.goreleaser: ENV GOMEMLIMIT=512MiB - main.go: blank import go.uber.org/automaxprocs (cgroup-aware GOMAXPROCS) - main.go: PPROF_PORT env var wires net/http/pprof on 127.0.0.1 only with full server timeouts - README.md: env-var docs + metric-label docs updated; cardinality note Test coverage push (per package): - main 51.2% -> 74.7% - cache 66.3% -> 93.7% - cache/redis 45.5% -> 98.2% - tracing 66.7% -> 72.9% - (cache/memory 91.6%, logging 91.9%, monitoring 77.6%, pkg/pools 100% unchanged) New test files: coverage_micro_test, coverage_extras_test, server_handlers_test, api_health_test, admin_dashboard_cluster_test, metrics_aggregator_test, concerns_test, cache/cache_coverage_test, cache/redis/redis_coverage_test, tracing/tracing_coverage_test. Bug fix: connection_resilience_test.go TestIntegratedHealthManagement.health_manager_startup was sync.Once-coupled to InitializeBackendHealth and panicked when another test (e.g. via parseConfig) had already triggered Once. Use NewBackendHealthManager directly.	2026-04-19 19:49:24 +01:00