mirror of
https://github.com/proffesor-for-testing/agentic-qe.git
synced 2026-09-19 08:45:47 +08:00
Dragan Spiridonov
ce1222a7c4
Real-time hook learning-capture was failing on this macOS Docker bind mount with `Failed to initialize UnifiedMemoryManager: database is locked`, and a brain-checkpoint hook was leaking deadlocked `brain export` subprocesses (4 found, ~1.5GB RSS). Root cause is SQLite concurrency on a WAL-hostile virtiofs mount, NOT a native-binary mismatch (better-sqlite3 loads fine). - brain-checkpoint.cjs: add AQE_DISABLE_BRAIN_CHECKPOINT env / marker-file kill switch (opt-in; shipped installs unaffected). The export spawns a native RVF writer that deadlocks in a futex and ignores the SIGTERM the timeout sends, so orphaned exporters piled up. Disabling stops the leak at the source. - aqe-hook.cjs: make the FATAL health-log fix-hint conditional. Only suggest `npm rebuild better-sqlite3` for real native markers; lock-contention failures now point at the actual cause instead of sending users down the wrong path. - feature-flags.ts: add the missing RUVECTOR_USE_RVF_PATTERN_STORE env override (the only RuVector flag lacking one) so an RVF FsyncFailed can fall back to the SQLite HNSW PatternStore. Activates after build. - scripts/aqe-db-backup.sh: consistent VACUUM INTO snapshots (safe with concurrent writers), integrity-verified before promotion, rotating + daily, to the host mount so they survive container loss. Restore round-trip tested. - devcontainer.json: auto-restart the backup loop (postStartCommand) and set bind-mount env mitigations (remoteEnv: AQE_DISABLE_WAL, RUVECTOR_USE_RVF_PATTERN_STORE=false, AQE_DISABLE_BRAIN_CHECKPOINT). - learning-config.json: clear the stale rebuildMode flag (no src reader; on since 2025-12). Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
66 lines
2.6 KiB
JSON
66 lines
2.6 KiB
JSON
{
|
|
"name": "Basic Development DevPod",
|
|
"image": "mcr.microsoft.com/devcontainers/base:debian12",
|
|
"runArgs": [
|
|
"--privileged",
|
|
"--memory=8g",
|
|
"--memory-swap=12g",
|
|
"--memory-reservation=4g",
|
|
"--cpus=3",
|
|
"--oom-kill-disable=false",
|
|
"--memory-swappiness=60"
|
|
],
|
|
"remoteUser": "vscode",
|
|
// Environment mitigations for this macOS virtiofs bind mount (see the
|
|
// postStartCommand backup loop and .agentic-qe/DISABLE_BRAIN_CHECKPOINT):
|
|
// - AQE_DISABLE_WAL: rollback-journal instead of WAL (WAL mmap corrupts here).
|
|
// - RUVECTOR_USE_RVF_PATTERN_STORE=false: RVF native init FsyncFails on this
|
|
// mount; fall back to SQLite HNSW. (Takes effect after `npm run build`.)
|
|
// - AQE_DISABLE_BRAIN_CHECKPOINT: stop the leaking `brain export` hook.
|
|
"remoteEnv": {
|
|
"AQE_DISABLE_WAL": "1",
|
|
"RUVECTOR_USE_RVF_PATTERN_STORE": "false",
|
|
"AQE_DISABLE_BRAIN_CHECKPOINT": "1"
|
|
},
|
|
"features": {
|
|
"ghcr.io/devcontainers/features/docker-in-docker:2": {},
|
|
"ghcr.io/devcontainers/features/node:1": {},
|
|
"ghcr.io/devcontainers/features/python:1": {}
|
|
},
|
|
"hostRequirements": {
|
|
"memory": "8gb",
|
|
"cpus": 3,
|
|
"storage": "64gb"
|
|
},
|
|
"customizations": {
|
|
"vscode": {
|
|
"extensions": [
|
|
"rooveterinaryinc.roo-cline",
|
|
"vsls-contrib.gistfs",
|
|
"github.copilot",
|
|
"anthropic.claude-code",
|
|
"codeflow-studio.claude-code-extension",
|
|
"suzuki0430.ccusage-vscode"
|
|
]
|
|
}
|
|
},
|
|
// node_modules lives in a named volume so the container installs its own
|
|
// (Linux) native binaries instead of seeing the host's (e.g. macOS) ones
|
|
// via the workspace bind mount. The host's node_modules is untouched.
|
|
// The volume is root-owned on first creation, hence the chown before npm ci.
|
|
"postCreateCommand": "sudo chown vscode:vscode node_modules && bash .devcontainer/install-tools.sh && npm ci",
|
|
// Auto-restart the verified learning-DB backup loop on every container start.
|
|
// memory.db lives on the macOS virtiofs bind mount (WAL-corruption prone), so
|
|
// scripts/aqe-db-backup.sh takes consistent, integrity-checked snapshots to
|
|
// .agentic-qe/backups/verified/ every 30 min — durable if the container dies.
|
|
// Idempotent: skips if a loop is already running.
|
|
"postStartCommand": "pgrep -f 'aqe-db-backup.sh loop' >/dev/null || (nohup bash scripts/aqe-db-backup.sh loop 1800 >/dev/null 2>&1 &)",
|
|
"shutdownAction": "none",
|
|
"mounts": [
|
|
"source=agentic-qe-node-modules,target=${containerWorkspaceFolder}/node_modules,type=volume"
|
|
],
|
|
"capAdd": [
|
|
"SYS_ADMIN",
|
|
"SYS_PTRACE"
|
|
]
|
|
} |