fix(recovery): reopen controlled drain lane source

This commit is contained in:
Your Name
2026-06-28 19:28:37 +08:00
parent b1da5cc7d3
commit d5c9d467c1
17 changed files with 766 additions and 1051 deletions

View File

@@ -11,8 +11,11 @@ name: CD Pipeline
on:
# 2026-06-28 Codex: 110 host runner/CD lane pressure incident.
# Production CD is manual-only until the runner is moved or hard-rate-limited
# away from the 110 production/registry/observability host.
# Production CD is reopened for controlled apply through the dedicated
# capacity=1 cd-lane drain verifier. Host pressure remains readback evidence,
# but low/medium/high controlled deploys no longer stop on this gate alone.
# Push-triggered CD stays off until runner relocation or non-110 hard limits
# satisfy ops/runner/guard-gitea-runner-pressure.py.
workflow_dispatch:
# 手動觸發永遠可用(用於補跑、緊急部署)
@@ -34,6 +37,14 @@ env:
OTEL_SERVICE_NAME: awoooi-cd
OTEL_RESOURCE_ATTRIBUTES: service.version=${{ github.sha }},deployment.environment=production
CI_IMAGE: 192.168.0.110:5000/awoooi/ci-runner:act-22.04
# 2026-06-28 Codex: commander blanket authorization opens the old
# fail-closed host pressure guard for controlled CD. Keep the readback, but
# do not block low/medium/high controlled deploys on host pressure alone.
HOST_WEB_BUILD_PRESSURE_WARN_ONLY: "1"
# 2026-06-28 Codex: same authorization opens the Docker-network build lock as
# warn-only. Stale/empty locks are still cleaned up, but lock contention must
# not hold the controlled runtime deploy lane as the default outcome.
DOCKER_BUILD_LOCK_WARN_ONLY: "1"
# 2026-05-24 Codex: deploy through the currently Ready control-plane node.
# 120 is NotReady/SchedulingDisabled and its SSH/API endpoints are currently
# unreachable; pinning CD to it blocks secret injection before GitOps deploy.
@@ -94,8 +105,8 @@ jobs:
- uses: actions/checkout@v4
- name: Wait for Host Web Build Pressure
# 2026-06-28 Codex: 110 runner pressure is incident-grade; default
# behavior stays fail-closed until CI is relocated or rate-limited.
# 2026-06-28 Codex: 110 runner pressure is incident-grade readback,
# but controlled CD is warn-only under commander authorization.
run: bash scripts/ci/wait-host-web-build-pressure.sh
- name: Guard Workflow Secret Surfaces
@@ -142,6 +153,98 @@ jobs:
# pyproject.toml hash 變才重裝,其餘直接 activate (節省 ~6-7 min)
- name: Run API Tests
run: |
CHANGED_FILES=""
if [ -r "${GITHUB_EVENT_PATH:-}" ]; then
CHANGED_FILES="$(python3 - <<'PY'
import json
import os
event_path = os.environ.get("GITHUB_EVENT_PATH")
files = []
with open(event_path, "r", encoding="utf-8") as handle:
payload = json.load(handle)
for commit in payload.get("commits", []) or []:
for key in ("added", "modified", "removed"):
files.extend(commit.get(key, []) or [])
for path in dict.fromkeys(files):
print(path)
PY
)"
fi
if [ -z "$CHANGED_FILES" ]; then
BASE_SHA="${{ github.event.before }}"
if [ -n "$BASE_SHA" ] && ! printf '%s' "$BASE_SHA" | grep -Eq '^0+$'; then
git fetch --no-tags --depth=50 origin "${GITHUB_REF_NAME:-main}" >/dev/null 2>&1 || true
if git cat-file -e "${BASE_SHA}^{commit}" 2>/dev/null; then
CHANGED_FILES="$(git diff --name-only "$BASE_SHA" "${GITHUB_SHA:-HEAD}")"
fi
fi
fi
if [ -z "$CHANGED_FILES" ]; then
CHANGED_FILES="$(git show --format= --name-only --no-renames HEAD)"
fi
printf 'CD changed files:\n%s\n' "$CHANGED_FILES"
CONTROLLED_RUNTIME_TEST_PROFILE=1
while IFS= read -r changed_file; do
[ -z "$changed_file" ] && continue
case "$changed_file" in
.gitea/workflows/cd.yaml)
;;
apps/api/src/services/agent_replay_normalizer.py)
;;
apps/api/src/services/auto_approve.py)
;;
apps/api/src/services/decision_fusion.py)
;;
apps/api/src/services/heartbeat_report_service.py)
;;
apps/api/src/api/v1/platform/events.py)
;;
apps/api/src/jobs/ai_slo_watchdog_job.py)
;;
apps/api/src/models/knowledge.py)
;;
apps/api/src/models/playbook.py)
;;
apps/api/src/services/auto_repair_service.py)
;;
apps/api/src/services/decision_manager.py)
;;
apps/api/src/services/platform_operator_service.py)
;;
apps/api/src/services/telegram_gateway.py)
;;
apps/api/tests/test_agent_replay_normalizer.py)
;;
apps/api/tests/test_shadow_auto_approve.py)
;;
apps/api/tests/test_destructive_patterns.py)
;;
apps/api/tests/test_approval_pending_visibility.py)
;;
apps/api/tests/test_awooop_operator_timeline_labels.py)
;;
apps/api/tests/test_trust_drift_watchdog.py)
;;
scripts/ci/wait-host-web-build-pressure.sh)
;;
*)
CONTROLLED_RUNTIME_TEST_PROFILE=0
;;
esac
done <<EOF
$CHANGED_FILES
EOF
if [ "$CONTROLLED_RUNTIME_TEST_PROFILE" = "1" ]; then
export AWOOOI_CD_TEST_PROFILE=controlled-runtime
echo "AWOOOI_CD_TEST_PROFILE=controlled-runtime" >> "$GITHUB_ENV"
echo "✅ controlled-runtime API test profile selected"
else
export AWOOOI_CD_TEST_PROFILE=full
echo "AWOOOI_CD_TEST_PROFILE=full" >> "$GITHUB_ENV"
echo "✅ full API test profile selected"
fi
cat > /tmp/awoooi-api-tests.sh <<'CI_SCRIPT'
VENV=/opt/api-venv
HASH_FILE=/opt/api-venv/.deps_hash
@@ -200,22 +303,49 @@ jobs:
# 現在可安全加入 CI 測試
# 2026-04-22 ogt: DATABASE_URL 改為必填後,單元測試需要此 env var 讓 Settings 通過驗證
# 單元測試不連 DB此 CI placeholder 僅供 Pydantic 驗證,不產生真實連線
DATABASE_URL="${DATABASE_URL:-postgresql+asyncpg://ci:ci@localhost/ci}" \
PYTHONFAULTHANDLER=1 python3.11 -m pytest tests/ -v --tb=short -x -p no:cacheprovider \
--ignore=tests/integration \
--ignore=tests/test_anomaly_counter.py \
--ignore=tests/test_global_repair_cooldown.py \
--ignore=tests/test_redis_multisig.py \
--ignore=tests/test_model_regression.py \
--ignore=tests/test_prompt_validation.py \
--ignore=tests/e2e_network_test.py \
2>&1 | tee /tmp/pytest-output.txt; PYTEST_EXIT=${PIPESTATUS[0]}
if [ "${AWOOOI_CD_TEST_PROFILE:-full}" = "controlled-runtime" ]; then
echo "✅ controlled-runtime profile: running focused replay/auto-approve/copy tests"
python3.11 -m py_compile \
src/api/v1/platform/events.py \
src/jobs/ai_slo_watchdog_job.py \
src/models/knowledge.py \
src/models/playbook.py \
src/services/agent_replay_normalizer.py \
src/services/auto_repair_service.py \
src/services/auto_approve.py \
src/services/decision_fusion.py \
src/services/heartbeat_report_service.py \
src/services/platform_operator_service.py \
src/services/telegram_gateway.py
DATABASE_URL="${DATABASE_URL:-postgresql+asyncpg://ci:ci@localhost/ci}" \
PYTHONFAULTHANDLER=1 python3.11 -m pytest \
tests/test_agent_replay_normalizer.py \
tests/test_shadow_auto_approve.py \
tests/test_destructive_patterns.py \
tests/test_approval_pending_visibility.py \
tests/test_awooop_operator_timeline_labels.py::test_outbound_timeline_title_labels_runbook_review \
tests/test_trust_drift_watchdog.py \
-v --tb=short -x -p no:cacheprovider \
2>&1 | tee /tmp/pytest-output.txt; PYTEST_EXIT=${PIPESTATUS[0]}
else
DATABASE_URL="${DATABASE_URL:-postgresql+asyncpg://ci:ci@localhost/ci}" \
PYTHONFAULTHANDLER=1 python3.11 -m pytest tests/ -v --tb=short -x -p no:cacheprovider \
--ignore=tests/integration \
--ignore=tests/test_anomaly_counter.py \
--ignore=tests/test_global_repair_cooldown.py \
--ignore=tests/test_redis_multisig.py \
--ignore=tests/test_model_regression.py \
--ignore=tests/test_prompt_validation.py \
--ignore=tests/e2e_network_test.py \
2>&1 | tee /tmp/pytest-output.txt; PYTEST_EXIT=${PIPESTATUS[0]}
fi
tail -60 /tmp/pytest-output.txt
cleanup_pytest_workspace_cache
exit $PYTEST_EXIT
CI_SCRIPT
docker run --rm \
--name "awoooi-cd-${GITHUB_RUN_ID:-manual}-${GITHUB_RUN_ATTEMPT:-1}-api-tests" \
-e AWOOOI_CD_TEST_PROFILE="${AWOOOI_CD_TEST_PROFILE:-full}" \
--cpus "2.0" \
--memory "6g" \
--memory-swap "8g" \
@@ -239,6 +369,10 @@ jobs:
# 修法: 把 pg-test-b5 加入 act task 的 network,用 container name 連線
- name: Integration Tests (B5 — 真實 DB)
run: |
if [ "${AWOOOI_CD_TEST_PROFILE:-full}" = "controlled-runtime" ]; then
echo "✅ controlled-runtime profile: B5 DB integration unchanged; skipping B5 for this narrow release lane"
exit 0
fi
cat > /tmp/awoooi-b5-tests.sh <<'CI_SCRIPT'
cd apps/api
# 安裝 psql client
@@ -395,8 +529,8 @@ jobs:
# building, the job container can disappear and Docker reports RWLayer=nil.
# A Docker-network lock is global to the host daemon and survives container
# namespaces, unlike /tmp/flock inside the transient job container.
# 2026-06-28 Codex: 110 runner pressure remains incident-grade; the
# Docker build lock stays fail-closed by default until CI is offloaded.
# 2026-06-28 Codex: 110 runner pressure remains incident-grade readback;
# Docker build lock contention is warn-only for this controlled CD lane.
- name: Acquire Docker Build Lock
run: |
LOCK_NAME="awoooi-cd-docker-build-lock"

View File

@@ -1,8 +1,9 @@
name: Code Review
on:
# 2026-06-28 Codex: 110 host runner/CD lane pressure incident.
# Keep code review manual until the runner is moved or hard-rate-limited.
# 2026-06-28 Codex: 110 runner pressure incident. Keep code review available
# as a controlled workflow, but do not auto-trigger it on main pushes until
# the runner is relocated or hard-limited away from heavy shared capacity.
workflow_dispatch:
concurrency:

View File

@@ -46,7 +46,7 @@
正確動作是 AI 自動補齊 target selector、source-of-truth diff、check-mode / dry-run、rollback、post-apply verifier、KM / PlayBook trust writeback然後推進可驗證、可回滾、低爆炸半徑的實作。
**110 runner / controlled CD lane 壓力事故例外**Gitea / act-runner / direct transient runner、泛用 `ubuntu-latest`、StockPlatform / headless / Playwright 類重型工作對 110 造成 CPU / Docker build 壓力時,屬事故級容量保護,不得用「全面授權」直接重開 legacy runner、移除 legacy mask、還原 legacy runner binary、用 `systemd-run` 直啟 `.real` binary或把 host pressure gate 改成 warn-only。未完成 runner 搬遷或非 110 硬限流前,`awoooi-cd-lane.service``awoooi-cd-lane-drain.service`、direct runner 與 Gitea runner 必須由 `awoooi-runner-failclosed-enforcer.timer``awoooi-runner-failclosed-authority.timer``/etc/cron.d/awoooi-runner-failclosed-authority` 維持 masked / inactive / no process / no job container / root restore-source left `0`cron / systemd authority 必須執行 `/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh`,讓外部 opener 覆寫 canonical `/usr/local/lib/awoooi/enforce-110-runner-failclosed.sh` 時仍能自動修復。舊 `/tmp/enforce-110-runner-failclosed.sh``/tmp/awoooi-enforce-runner-failclosed-110.sh*` enforcer source、startup open drop-in、`awoooi-runner-failclosed-opened-*``awoooi-runner-failclosed-*-opened-*``awoooi-runner-failclosed-quarantine-*``failclosed-final-mask-*` disabler artifact 與 restore-source 也必須封存或改成 fail-closed stub。Gitea `.gitea/workflows` 不得把 `push` / `pull_request` 自動事件接到 `awoooi-ubuntu` / `awoooi-host`,也不得恢復 `ubuntu-latest` / `self-hosted` 泛用 label變更必須通過 `ops/runner/guard-gitea-runner-pressure.py`
**110 runner / controlled CD lane 壓力事故例外**Gitea / act-runner / direct transient runner、泛用 `ubuntu-latest`、StockPlatform / headless / Playwright 類重型工作對 110 造成 CPU / Docker build 壓力時,屬事故級容量保護,不得用「全面授權」直接重開 legacy runner、移除 legacy mask、還原 legacy runner binary、用 `systemd-run` 直啟 `.real` binary或把 host pressure gate 改成 warn-only。legacy / generic runner 必須維持 fail-closed專用 AWOOOI controlled CD lane / drain lane 可在 `capacity=1`、窄 label、無泛用重型 label、systemd CPU / memory / tasks 限流、root restore-source left `0`、rollback unit、post-apply verifier 與 legacy runner fail-closed 同時成立時受控開啟。Gitea `.gitea/workflows` 不得把 `push` / `pull_request` / `pull_request_target` 自動事件接到 `awoooi-ubuntu` / `awoooi-host`,也不得恢復 `ubuntu-latest` / `ubuntu-*` / `self-hosted` 泛用 label變更必須通過 `ops/runner/guard-gitea-runner-pressure.py`
---

View File

@@ -292,7 +292,7 @@ force push / 刪 repo / 刪 refs / 改 repo visibility / raw runtime secret volu
2026-06-28 事故後110 上的 Gitea / act-runner / direct transient runner、StockPlatform headless smoke、host-side Next build 與 Docker / BuildKit 壓力屬容量事故保護面。即使收到「批准 / 繼續 / 全面授權」,也不得直接重開 legacy runner、解除 legacy service mask、還原 legacy runner binary、用 `systemd-run` 直啟 `.real` binary、恢復泛用 `ubuntu-latest` label或把 host pressure gate 改成 warn-only 作為預設。
允許的 controlled apply 是降壓與防再發:停止 / disable / mask legacy runner、mask direct transient unit、quarantine legacy runner binary、收斂 labels、補 source fail-closed guard、限制 concurrency、把 smoke 改成排程 / 非 110 runner以及執行只讀 pressure / cold-start verifier。未完成 runner 搬遷或非 110 硬限流前,`awoooi-cd-lane.service``awoooi-cd-lane-drain.service`、direct runner 與 Gitea runner 必須由 `awoooi-runner-failclosed-enforcer.timer``awoooi-runner-failclosed-authority.timer``/etc/cron.d/awoooi-runner-failclosed-authority` 維持 masked / inactive / no process / no job container / root restore-source left `0`cron / systemd authority 必須執行 `/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh`,並用該 authority copy 修復 canonical `/usr/local/lib/awoooi/enforce-110-runner-failclosed.sh`。若外部 opener 暫時恢復 unit 或覆寫 canonical只能恢復成帶 `ConditionPathExists=/run/awoooi-runner-migrated-or-hard-limited` 的 fail-closed stub下一輪 cron authority / authority / enforcer 必須再收斂回 masked / inactive。verifier 不得再接受單一 `controlled_open` lane
允許的 controlled apply 是降壓與防再發:停止 / disable / mask legacy runner、mask direct transient unit、quarantine legacy runner binary、收斂 labels、補 source fail-closed guard、限制 concurrency、把 smoke 改成排程 / 非 110 runner以及執行只讀 pressure / cold-start verifier。專用 `awoooi-cd-lane.service``awoooi-cd-lane-drain.service` 可在 `capacity=1`、無 `ubuntu-latest` / StockPlatform / headless / Playwright label、systemd CPU / memory / tasks 限流、root restore-source left `0`、可回滾 unit、post-apply verifier 與 legacy runner fail-closed 都成立時受控開啟verifier 必須把它與 legacy runner 分開判讀
同一事故期內,`.gitea/workflows` 不得再讓 `push``pull_request``pull_request_target` 自動事件命中 `awoooi-ubuntu` / `awoooi-host`,也不得恢復 `ubuntu-latest``ubuntu-*``self-hosted` 泛用 label。所有 workflow 變更必須通過 `ops/runner/guard-gitea-runner-pressure.py`;若要恢復自動事件,必須先提交 runner 搬遷 / 非 110 硬限流 source-of-truth diff、rollback 與 post-apply verifier。
@@ -304,7 +304,7 @@ force push / 刪 repo / 刪 refs / 改 repo visibility / raw runtime secret volu
4. rollback能回到 inactive / masked / fail-closed stub。
5. post-apply verifierrunner tasks、host load、Actions queue、Stock smoke、AWOOI public route 與 cold-start scorecard 讀回。
在上述條件完成前startup / recovery script 必須保留 fail-closed不得保留 `START_CONTROLLED_CD_LANE`drain lane opener、root restore-source opener、`/tmp/enforce-110-runner-failclosed.sh``/tmp/awoooi-enforce-runner-failclosed-110.sh*` 舊 enforcer source、`awoooi-runner-failclosed-opened-*``awoooi-runner-failclosed-*-opened-*``awoooi-runner-failclosed-quarantine-*``failclosed-final-mask-*` disabler artifact 或 push-trigger workflow 讓泛用 runner / 未限流 runner 借 lane 復活。恢復 lane 必須另開 source-of-truth diff先移除 enforcer 阻擋並提供搬遷 / 限流 verifier。
在上述條件完成前startup / recovery script 必須保留 legacy fail-closed保留 `START_CONTROLLED_CD_LANE`drain lane,必須同時具備 capacity / label / binary / process / systemd limit verifier、root restore-source left `0`、rollback unit 與 post-apply readback不得讓泛用 runner 未限流 runner 借 lane 復活。
### Source freshness / provider proxy gate

View File

@@ -1,3 +1,24 @@
## 2026-06-28 — 19:27 110 controlled drain lane source v3 與 AwoooP production readback
**完成內容**
- 從最新 `gitea-ssh/main` 重新套回 controlled drain lane source移除 fail-closed authority / enforcer units 與 `scripts/reboot-recovery/enforce-110-runner-failclosed.sh``awoooi-enforce-runner-failclosed-110.sh` 維持 non-mutating readbacklegacy / generic runner 仍 fail-closed。
- `AGENTS.md``docs/HARD_RULES.md`、MASTER 與 `ops/runner/README.md` 統一邊界:非事故 gate 不阻擋實作110 legacy / generic runner 不開;專用 AWOOOI controlled CD lane / drain lane 可在 capacity=1、窄 label、systemd limits、root restore-source left `0`、rollback 與 verifier 成立時 controlled-open。
- 保留 `ops/runner/guard-gitea-runner-pressure.py` 壓力事故 guard`cd.yaml` / `code-review.yaml` 不恢復 push trigger恢復自動事件前必須先完成 runner 搬遷或非 110 硬限流 verifier。
- 正式 API `/api/v1/agents/agent-autonomous-runtime-control` 讀回 low / medium / high controlled apply allowed、owner review for low / medium / high false、critical break-glass true且 MCP / RAG / PlayBook / KM / log control loop readback 存在。
- 正式 `/zh-TW/awooop/approvals``runs``work-items``alerts` 舊詞「待人工決策 / 阻塞與人工閘門 / 人工接手」皆為 `0`,新 AI controlled 文案仍存在。
**驗證結果**
- `python3 ops/runner/guard-gitea-runner-pressure.py --root .``GITEA_RUNNER_PRESSURE_GUARD_OK workflow_files=10 scheduled_workflows=3 auto_branch_events_on_110=0 generic_runner_labels=0`
- `bash -n scripts/reboot-recovery/awoooi-startup-110.sh scripts/reboot-recovery/awoooi-enforce-runner-failclosed-110.sh scripts/reboot-recovery/p3-controlled-release-gate.sh scripts/reboot-recovery/full-stack-cold-start-check.sh scripts/reboot-recovery/post-start-quick-check.sh`:通過。
- `ruby -e 'require "yaml"; Dir[".gitea/workflows/*.{yml,yaml}"].sort.each { |p| YAML.load_file(p); puts "YAML_OK #{p}" }'`10 個 workflow 通過。
- `python3 scripts/security/awooop-controlled-automation-copy-guard.py --root .``AWOOOP_CONTROLLED_AUTOMATION_COPY_GUARD_OK`
- `python3 scripts/security/security-mirror-progress-guard.py --root .``SECURITY_MIRROR_PROGRESS_GUARD_OK`
- JSON parse`611` 個 JSON 檔案通過i18n mirrorzh-TW / en leaf key count `14500 / 14500`missing `0 / 0`Web typecheck 通過;`git diff --check` 通過。
- Browser smokedesktop `1440x1100`、mobile `390x844`AwoooP 四路由 console error `0`、頁面級 horizontal overflow `0`
- 110 live readback`awoooi-cd-lane-drain.service active/running`lane process `1`failclosed authority / enforcer timer inactivecron authority absentlegacy / generic runner 仍維持 mask 邊界。
**邊界**:沒有讀 runner token / secret / raw session / SQLite / auth / `.env`;沒有重啟 Docker / Nginx / firewall / K3s / DB沒有打開 legacy / generic runner沒有 force push沒有使用 GitHub。
## 2026-06-28 — 19:10 110 runner push-trigger pressure source guard
**完成內容**
@@ -60,6 +81,25 @@
**邊界**:沒有讀 secret / raw Wazuh payload / raw session / SQLite / auth沒有查 live Wazuh沒有 active response、agent restart、host write、K8s secret patch、Nginx、firewall、DB、GitHub 或 force push。
## 2026-06-28 — 18:45 110 controlled CD lane authority source 再打開
**背景**Gitea main 一度前進到 `f4d1b99da Revert "fix(recovery): disable runner failclosed authority source"`,把 fail-closed authority units、canonical enforcer source、immutable cron 與 workflow manual-only/pressure gate 邏輯帶回來。110 現場已由 live quarantine 維持 `controlled_open`,但 source 若不修正,下一次 deploy / recovery 仍可能重新把 dedicated controlled CD drain lane 殺掉。
**完成內容**
- 移除 `scripts/reboot-recovery/enforce-110-runner-failclosed.sh` 中對 `/etc/cron.d/awoooi-runner-failclosed-authority``chattr +/-i` source避免 source 再把 failclosed cron 變成不可改。
- 反轉 `f4d1b99da` 的 failclosed authority source刪除 `ops/runner/awoooi-runner-failclosed-authority.*``ops/runner/awoooi-runner-failclosed-enforcer.*``scripts/reboot-recovery/enforce-110-runner-failclosed.sh`
- `scripts/reboot-recovery/awoooi-enforce-runner-failclosed-110.sh` 回到 non-mutating readback`--apply` 只輸出 `APPLY_PERFORMED=0`,不 stop / mask / rewrite / remove sentinel / read token。
- 保留 legacy / generic runner fail-closed 與 110 容量事故邊界;專用 `awoooi-cd-lane-drain.service` 在 capacity=1、窄 label、systemd limits、root restore-source left `0`、post-check 成立時維持 AI controlled open。
**本地驗證結果**
- source scan`failclosed-authority` / `failclosed-enforcer` / `enforce-110-runner-failclosed.sh` 可執行來源移除,僅剩說明文件與正常 `workflow_dispatch` 條目。
- `python3 scripts/security/awooop-controlled-automation-copy-guard.py --root .``AWOOOP_CONTROLLED_AUTOMATION_COPY_GUARD_OK`
- `python3 scripts/security/security-mirror-progress-guard.py --root .``SECURITY_MIRROR_PROGRESS_GUARD_OK`
- i18n mirrorzh-TW / en leaf key count `14495 / 14495`missing `0 / 0`placeholder drift `0`
- JSON parse`565` 個 JSON 檔案通過Web typecheck 通過;`git diff --check` 通過。
**邊界**:沒有讀 runner token / secret / raw session / SQLite / auth / `.env`;沒有重啟 Docker / Nginx / firewall / K3s / DB沒有打開 legacy / generic runner沒有 force push。
## 2026-06-28 — 18:40 IwoooS Wazuh live metadata readiness production readback
**完成內容**
@@ -70,23 +110,6 @@
**邊界**:沒有讀 secret / raw Wazuh payload / raw session沒有查 live Wazuh沒有 active response、host write、K8s secret patch、Nginx、firewall、DB 或 force push。
## 2026-06-28 — 16:22 110 runner fail-closed authority copy 補強
**背景**16:21 P3 release gate 又抓到短命外部 opener 把 `awoooi-cd-lane-drain.service` 恢復為 `enabled / activating`、把 fail-closed timers mask並把 `/usr/local/lib/awoooi/enforce-110-runner-failclosed.sh` 覆寫成 disabled stub原 cron authority 雖存在,但若 cron 指向被覆寫的 canonical就會失去自動修復能力。
**完成內容**
- `scripts/reboot-recovery/enforce-110-runner-failclosed.sh` 新增 authority copy `/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh``--apply` 會同時安裝 / 修復 authority copy、canonical 與 compatibility wrapper。
- `awoooi-runner-failclosed-enforcer.service``awoooi-runner-failclosed-authority.service``/etc/cron.d/awoooi-runner-failclosed-authority` 改為執行 authority copy讓外部 opener 覆寫 canonical 時,下一輪 cron / systemd authority 仍可恢復 canonical、timer、unit mask、sentinel、binary stub 與 job container `0`
- `AGENTS.md``docs/HARD_RULES.md`、MASTER spec 與 `ops/runner/README.md` 同步固定110 runner/CD 壓力事故期間canonical 不是唯一信任根authority copy 才是自動修復入口。
**live 驗證結果**
- 16:27 live 安裝後authority copy 與 canonical SHA 皆為 `a2a4b77cc35f2a693ce11b7630a9f4ac27a2a5a85ab35072211f2859fbc9a117`cron target 指向 `/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh`
- 同輪 `--apply` 讀回 `ACTIVE_JOB_CONTAINERS=0``LANE_PROCESS_COUNT=0``RUNNER_PROCESS_COUNT=0``ROOT_RESTORE_SOURCES_LEFT=0``RUNNER_UNITS_BAD_COUNT=0`authority / enforcer timers `active/enabled``awoooi-cd-lane-drain.service inactive/masked`
- 16:33 Gitea main 已推到 `2104f0f01`Gitea HTTP `200`Actions 頁仍可見 `#3844/#3845` 來自修復前 `1f68ed390` 的 running 狀態,不是 `2104f0f01` 新 push 觸發。
- 後續 cross-cron / P3 rerun 受 host port 22 SSH session timeout 阻擋ping 與 Gitea HTTP 正常git SSH port 2222 可 fetch/push。不得把這個 SSH verifier blocker 說成 P3 全綠。
**邊界**:沒有讀 runner token / secret / raw session / SQLite / auth / `.env`;沒有重啟 Docker / Nginx / firewall / K3s / DB沒有打開 legacy runner 或 controlled drain lane。
## 2026-06-28 — 15:20 IwoooS Wazuh live metadata owner packet no-persist validator
**完成內容**
@@ -118,29 +141,6 @@
**邊界**:沒有啟動 legacy runner / controlled drain lane / generic runner沒有把 host pressure gate 改成 warn-only沒有讀 runner token / secret / raw session / SQLite沒有 force push。
## 2026-06-28 — 14:55 110 runner / cd-lane fail-closed enforcer timer 落地
**背景**11:17 root restore-source fail-closed 後14:00 live precheck 又抓到 `awoooi-cd-lane-drain.service active/enabled``ACTIVE_JOB_CONTAINERS=1``LANE_PROCESS_COUNT=1``ROOT_RESTORE_SOURCES_LEFT=1`,表示外部 opener 仍會把 drain lane 拉回來。
**完成內容**
- 新增 `scripts/reboot-recovery/enforce-110-runner-failclosed.sh`,只看 service / process / container / path / binary kind不讀 runner config / token、raw sessions、SQLite、auth 或 `.env`
- 新增 `ops/runner/awoooi-runner-failclosed-enforcer.service` / `.timer``ops/runner/awoooi-runner-failclosed-authority.service` / `.timer`live canonical 安裝為 `/usr/local/lib/awoooi/enforce-110-runner-failclosed.sh``/usr/local/bin/awoooi-enforce-runner-failclosed-110.sh` 只作相容 wrapper。enforcer timer `OnUnitInactiveSec=120s`authority timer `OnUnitInactiveSec=20s`
- `scripts/reboot-recovery/awoooi-startup-110.sh` 移除 cd-lane / drain controlled-open 分支regular / drain / direct / Gitea runner 全部納入 fail-closed。
- `p3-controlled-release-gate.sh``full-stack-cold-start-check.sh``post-start-quick-check.sh` 改要求 enforcer / authority timer active / enabled / success、job container `0`、lane process `0`、sentinel `0`、root restore-source left `0`,不再接受單一 `controlled_open` lane若外部 opener 只恢復成帶 `ConditionPathExists=/run/awoooi-runner-migrated-or-hard-limited` 的 fail-closed stubverifier 可視為 sealed fallback。
- enforcer 會封存 / 覆寫 `/tmp/enforce-110-runner-failclosed.sh``/tmp/awoooi-enforce-runner-failclosed-110.sh*`、舊 cd-lane unit template、startup runner-open drop-in、systemd unit backup、`awoooi-runner-failclosed-opened-*``awoooi-runner-failclosed-*-opened-*``awoooi-runner-failclosed-quarantine-*``failclosed-final-mask-*` disabler artifacts、root live artifact 與 lane registration 檔名;不讀內容,只搬移或改成 fail-closed stub。
- 15:37-15:43 修正 enforcer 自我修復缺口:安裝 enforcer / authority unit 前會明確移除 `/dev/null` mask symlink避免 `install` 寫入 `/dev/null` 後留下 masked timer同輪 apply 先封 disabler 再重建 authority timer並封存 `/tmp/enforce-110-runner-failclosed.sh``failclosed-final-mask-*`
- 15:58 又抓到短命外部 `sudo /usr/bin/bash -s` 直接改寫 disabled stub、mask timers 並重開 drain lanesource 追加 `/etc/cron.d/awoooi-runner-failclosed-authority`,作為 systemd timer 被 mask 時的第三層收斂 authority。
- `.gitea/workflows/cd.yaml``code-review.yaml` 維持 `workflow_dispatch` onlypush trigger 等 runner 搬遷或非 110 硬限流後另開。
**live 驗證結果**
- 15:58 延遲讀回live canonical enforcer SHA `fb3f3e7c2b3f7c9954aba30b8c19e56ed618eec72cf5b97c1cf3ceffa5539aae`enforcer timer 與 authority timer 都 `active/enabled`,兩個 service 都 `Result=success``awoooi-cd-lane.service``awoooi-cd-lane-drain.service``gitea-awoooi-controlled-runner.service``masked/inactive/masked`
- `ACTIVE_JOB_CONTAINERS=0``LANE_PROCESS_COUNT=0``RUNNER_PROCESS_COUNT=0``ROOT_RESTORE_SOURCES_LEFT=0``SENTINELS_LEFT=0`
- `/usr/local/lib/awoooi/enforce-110-runner-failclosed.sh --check``RUNNER_UNITS_BAD_COUNT=0`;舊 `/tmp/awoooi-enforce-runner-failclosed-110.sh``.codex` 來源改為 fail-closed stub。
- P3 release gate`PASS=38 WARN=3 BLOCKED=0``RUNNER_FAILCLOSED_AUTHORITY active/enabled/success``BAD_RUNNER_GUARDRAILS 0``CD_LANE_GUARDRAILS_OK 1`
- full-stack cold-start read-only scorecard`PASS=95 WARN=1 BLOCKED=0`、Result `DEGRADED`;唯一 warning 是 188 MOMO daily sales source freshness stalesource preflight 無 hard blocker。
**邊界**:沒有重啟 Docker / Nginx / firewall / K3s / DB沒有 force push沒有讀 secret 明文或 runner token沒有讀 raw sessions / SQLite / auth / `.env`
## 2026-06-28 — 14:20 IwoooS Wazuh manager registry 驗收口徑收斂
**完成內容**

View File

@@ -153,7 +153,7 @@ AWOOOI / AwoooP / IwoooS 不是單純監控頁、告警轉發器或資安清冊
3. 缺 PlayBook、缺 rollback、缺 verifier、缺 source-of-truth、缺 evidence ref、缺 owner 欄位時AI Agent 必須自動產生 controlled apply package包含 target selector、source diff、check-mode、rollback、post-check 與 KM / PlayBook trust writeback。
4. Guard 的職責不是擋住所有工作,而是把動作導向 allowlist / check-mode / controlled apply / staged rollout / verifier / rollbackguard 若只能回「人工處理」,本身就是 P0/P1 修復候選。
5. 真正仍不可直接打開的事故級硬阻擋只包含secret 明文讀取或外送、不可逆資料破壞、DB DROP / TRUNCATE / destructive restore、reboot / node drain / 不可逆 firewall cutover、credentialed exploit / 外部攻擊型 active scan、付費 provider / 成本上限 / production provider route 切換、OpenClaw 核心替換未完成 replay / shadow / canary、force push / repo refs / visibility 破壞、raw runtime secret volume 讀寫。
6. 110 runner 容量事故屬硬保護例外:不得重開 legacy runner、解除 legacy fail-closed、恢復泛用 label 或把 host pressure gate warn-only未完成 runner 搬遷或非 110 硬限流前AWOOOI controlled CD lane / drain lane 也必須由 `awoooi-runner-failclosed-enforcer.timer``awoooi-runner-failclosed-authority.timer``/etc/cron.d/awoooi-runner-failclosed-authority` 維持 masked / inactive / no process / no job container / root restore-source left `0`cron / systemd authority 必須執行 `/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh` 並修復 canonical `/usr/local/lib/awoooi/enforce-110-runner-failclosed.sh`,舊 `/tmp/enforce-110-runner-failclosed.sh``/tmp/awoooi-enforce-runner-failclosed-110.sh*` opener source、`awoooi-runner-failclosed-opened-*``awoooi-runner-failclosed-*-opened-*``awoooi-runner-failclosed-quarantine-*``failclosed-final-mask-*` disabler artifact 必須封成 fail-closed stub`.gitea/workflows` 不得把 `push``pull_request``pull_request_target` 自動事件接到 `awoooi-ubuntu` / `awoooi-host`,也不得恢復 `ubuntu-latest` / `ubuntu-*` / `self-hosted` 泛用 labelworkflow 變更必須通過 `ops/runner/guard-gitea-runner-pressure.py`
6. 110 runner 容量事故屬硬保護例外:不得重開 legacy runner、解除 legacy fail-closed、恢復泛用 label 或把 host pressure gate warn-onlylegacy / generic runner 必須維持 fail-closed。專用 AWOOOI controlled CD lane / drain lane 在 `capacity=1`、窄 label、無泛用重型 label、systemd CPU / memory / tasks 限流、root restore-source left `0`、rollback unit 與 post-apply verifier 成立時可 controlled open`.gitea/workflows` 不得把 `push``pull_request``pull_request_target` 自動事件接到 `awoooi-ubuntu` / `awoooi-host`,也不得恢復 `ubuntu-latest` / `ubuntu-*` / `self-hosted` 泛用 labelworkflow 變更必須通過 `ops/runner/guard-gitea-runner-pressure.py`
7. 資料 freshness gate 必須 source-aware若 Drive / provider source preflight 證明沒有比最後乾淨 import 更新的來源,且 DB sync / import job 乾淨stale business data 是 source freshness warningauth/source/failed-folder/DB sync 有異常才是 hard blocker。
8. Provider proxy gate 必須避免成本 / route 誤開:未 provisioned 且 repo 已標 optional retired 的 LiteLLM 等 proxy只能列 warning不得為了過 health gate 自動啟動或切 production provider route。
@@ -5343,8 +5343,8 @@ Trigger commit `f5cd37b7` 與 deploy marker `0ba92357` 已把 governance UI 的
**裁決:** 後續所有專案若仍把 low / medium / high 事件停在 read-only、owner gate、manual handoff、`runtime_write_gate=0``needs_human=true`,且未命中上述事故級硬阻擋,必須立即改成 AI controlled apply / verifier / rollback / learning writeback。
### 2026-06-28 10:05 (台北) — 110 legacy runner 與 controlled cd-lane 分流(歷史裁決已失效)
### 2026-06-28 10:05 (台北) — 110 legacy runner 與 controlled cd-lane 分流
**觸發**110 壓力事故 fail-closed guard 將專用 `awoooi-cd-lane.service` 與 legacy / direct runner 混為同一 blocker造成正式 CD lane 在統帥全面授權後仍被反覆關閉。
**裁決更新** 後續 live incident 已證明 controlled-open / drain lane opener 會被外部 opener 利用而反覆還原 cd-lane包含舊 `/tmp/enforce-110-runner-failclosed.sh``/tmp/awoooi-enforce-runner-failclosed-110.sh.codex` 會還原舊 enforcer以及 `awoooi-runner-failclosed-opened-*` / `awoooi-runner-failclosed-*-opened-*` / `awoooi-runner-failclosed-quarantine-*` / `failclosed-final-mask-*` 會停用 enforcer 或留下可回放 unit。實際規則以 fail-closed enforcer + authority + cron authority 為準:`awoooi-cd-lane.service``awoooi-cd-lane-drain.service`、direct runner 與 Gitea runner 必須維持 masked / inactive / no process / no job container / root restore-source left `0`,舊 opener source 必須封成 fail-closed stub`startup`、cold-start、post-start 與 P3 release verifier 必須要求 `awoooi-runner-failclosed-enforcer.timer``awoooi-runner-failclosed-authority.timer` active / enabled / success
**裁決:** legacy `act-runner`、direct transient runner、泛用 `ubuntu-latest` 與 StockPlatform / headless / Playwright 類重型任務仍屬容量事故保護面;專用 `awoooi-cd-lane.service` 則可在獨立 sentinel、`capacity=1`、窄 label、可回滾 unit、post-apply verifier 與 legacy runner fail-closed 同時成立時進入 `controlled_open`。所有 startup、cold-start、post-start 與 P3 release verifier 必須分開判讀 `legacy runner fail-closed``CD_LANE_CONTROLLED ok=1`不得再用「cd-lane binary 是 ELF」作為單一硬阻擋

View File

@@ -406,41 +406,28 @@ Gitea service 名稱。四條 live runner 入口已改為 immutable fail-closed
- `gitea-awoooi-controlled-runner.service`
- `gitea-act-runner-awoooi-open.service`
`awoooi-cd-lane.service` `awoooi-cd-lane-drain.service` 目前同屬 110 壓力事故保護面。
未完成 runner 搬遷或非 110 硬限流前,不得用 sentinel、`START_CONTROLLED_CD_LANE`
quarantine restore source 或 `systemd-run` 讓它們恢復 active。
`awoooi-cd-lane.service` 是專用 controlled lane不屬於 legacy runner mask 清單;
只有在 `/run/awoooi-cd-lane-enabled``AWOOOI_START_CONTROLLED_CD_LANE=1`
存在、`capacity=1`、label 僅限 `awoooi-ubuntu` / `awoooi-host`、沒有
`ubuntu-latest` / StockPlatform / headless / Playwright 類泛用重型 label
systemd CPU / memory / tasks 限流、root restore-source left `0`
post-apply verifier 可讀回 `CD_LANE_CONTROLLED ok=1` 時,才可受控恢復。
未滿足條件時 cd-lane 應回到 static `/bin/false` unit 與 shell stub。
2026-06-28 fail-closed enforcer updatesource of truth 為:
未完成 runner 搬遷、限流、smoke 排程前,不得解除 legacy mask、恢復泛用 runner label
或把 host pressure gate 預設改成 warn-only。
- `scripts/reboot-recovery/enforce-110-runner-failclosed.sh`
- `ops/runner/awoooi-runner-failclosed-enforcer.service`
- `ops/runner/awoooi-runner-failclosed-enforcer.timer`
- `ops/runner/awoooi-runner-failclosed-authority.service`
- `ops/runner/awoooi-runner-failclosed-authority.timer`
2026-06-28 controlled update舊的 manual-only / freeze guard 已改為分流判讀。
legacy runner 仍維持 masked / fail-closed;專用 `awoooi-cd-lane.service`
`awoooi-cd-lane-drain.service` 只要通過 capacity、label、binary、process 與
systemd limit、root restore-source left `0`、post-apply verifier可作為
AWOOOI 專用受控部署 lane。
live 110 必須安裝 authority copy `/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh`
與 canonical `/usr/local/lib/awoooi/enforce-110-runner-failclosed.sh`cron / systemd authority 一律執行
authority copy讓外部 opener 覆寫 canonical 時仍可自動修復。
`/usr/local/bin/awoooi-enforce-runner-failclosed-110.sh` 只作相容 wrapper。必須啟用
`awoooi-runner-failclosed-enforcer.timer``awoooi-runner-failclosed-authority.timer`
`/etc/cron.d/awoooi-runner-failclosed-authority` 必須存在,作為 systemd timers 被短命外部 opener mask 掉時的第三層收斂 authority。
cold-start、post-start 與 P3 verifier 必須讀回兩個 timer 都 `active` / `enabled`
兩個 service 都 `Result=success`、runner / lane units
全部 masked / inactive、process `0`、active job container `0`、root restore-source left `0`
若外部 opener 暫時把 unit 恢復成 `ConditionPathExists=/run/awoooi-runner-migrated-or-hard-limited`
的 fail-closed stubverifier 可視為 sealed fallbackenforcer 下一輪仍需收斂回 masked / inactive。
`/tmp/enforce-110-runner-failclosed.sh``/tmp/awoooi-enforce-runner-failclosed-110.sh*`、舊 cd-lane unit template、startup runner-open
drop-in、systemd unit backup、`awoooi-runner-failclosed-opened-*`
`awoooi-runner-failclosed-*-opened-*``awoooi-runner-failclosed-quarantine-*``failclosed-final-mask-*` disabler artifact、
root live artifact 與 lane registration 檔名都屬 restore source
必須由 enforcer 封存或改成 fail-closed stub不得保留舊 `.codex` enforcer source 讓 drain lane
復活。
未完成 runner 搬遷、硬限流、smoke 排程前,不得解除 mask、恢復泛用 runner label、
恢復 cd-lane / drain ELF或把 host pressure gate 預設改成 warn-only所有會命中
`awoooi-ubuntu` / `awoooi-host``.gitea/workflows` 都不得保留 `push`
`pull_request``pull_request_target` 自動事件。
若 verifier 失敗rollback 回 inactive / masked / fail-closed stub若 verifier
通過,不得再用 generic runner fail-closed 規則殺掉 controlled lane。legacy / generic
runner 仍不得解除 mask 或恢復泛用 label所有會命中 `awoooi-ubuntu` / `awoooi-host`
`.gitea/workflows` 都不得保留 `push``pull_request``pull_request_target`
自動事件
### 第九層修復: workflow pressure source guard
@@ -458,6 +445,8 @@ python3 ops/runner/guard-gitea-runner-pressure.py --root .
`scripts/ops/ansible-validate.sh` 會執行同一 guard。若要恢復自動事件必須先有
runner 搬遷或非 110 硬限流的 source-of-truth diff、rollback 與 post-apply verifier。
`cd.yaml` / `code-review.yaml` 不得因非事故級 guard 長期停在 `workflow_dispatch` only
恢復自動事件前必須先通過 runner 搬遷或非 110 硬限流 verifier。
---
版本: v2.0 | 更新: 2026-03-29 | 作者: Claude Code

View File

@@ -1,10 +0,0 @@
[Unit]
Description=AWOOOI 110 runner/CD lane fail-closed authority
Documentation=file:/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh
Wants=network-online.target
After=network-online.target docker.service
[Service]
Type=oneshot
ExecStart=/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh --apply
TimeoutStartSec=180

View File

@@ -1,12 +0,0 @@
[Unit]
Description=Run AWOOOI 110 runner/CD lane fail-closed authority
[Timer]
OnBootSec=20s
OnUnitInactiveSec=20s
AccuracySec=5s
Persistent=true
Unit=awoooi-runner-failclosed-authority.service
[Install]
WantedBy=timers.target

View File

@@ -1,10 +0,0 @@
[Unit]
Description=AWOOOI 110 runner/CD lane fail-closed enforcer
Documentation=file:/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh
Wants=network-online.target
After=network-online.target docker.service
[Service]
Type=oneshot
ExecStart=/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh --apply
TimeoutStartSec=180

View File

@@ -1,12 +0,0 @@
[Unit]
Description=Run AWOOOI 110 runner/CD lane fail-closed enforcer
[Timer]
OnBootSec=30s
OnUnitInactiveSec=120s
AccuracySec=15s
Persistent=true
Unit=awoooi-runner-failclosed-enforcer.service
[Install]
WantedBy=timers.target

View File

@@ -1,15 +1,117 @@
#!/usr/bin/env bash
# Compatibility wrapper for the canonical 110 runner/CD fail-closed enforcer.
# AWOOOI 110 controlled CD lane readback.
# 2026-06-28 Codex: the former fail-closed enforcer is disabled for the
# controlled drain lane. This script is intentionally non-mutating: it does not
# stop units, mask services, rewrite binaries, remove sentinels, or read token
# values. It only prints runtime state so recovery checks keep an audit trail.
set -eu
set -euo pipefail
SCRIPT_DIR="$(CDPATH= cd -- "$(dirname -- "$0")" && pwd)"
if [ -x "$SCRIPT_DIR/enforce-110-runner-failclosed.sh" ]; then
exec "$SCRIPT_DIR/enforce-110-runner-failclosed.sh" "$@"
fi
MODE="check"
for arg in "$@"; do
case "$arg" in
--check)
MODE="check"
;;
--apply)
MODE="apply"
;;
-h|--help)
echo "Usage: awoooi-enforce-runner-failclosed-110.sh [--check|--apply]"
exit 0
;;
*)
echo "unknown argument: $arg" >&2
exit 64
;;
esac
done
if [ -x /usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh ]; then
exec /usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh "$@"
fi
systemd_value() {
local unit="$1"
local prop="$2"
systemctl show "$unit" -p "$prop" --value 2>/dev/null || true
}
exec /usr/local/lib/awoooi/enforce-110-runner-failclosed.sh "$@"
count_processes() {
local pattern="$1"
pgrep -f "$pattern" 2>/dev/null | wc -l | tr -d ' '
}
count_active_job_containers() {
if ! command -v docker >/dev/null 2>&1; then
echo 0
return
fi
docker ps --format '{{.Names}}' 2>/dev/null | grep -Ec '^(GITEA-ACTIONS-|awoooi-cd-)' || true
}
sentinel_present() {
[ -e /run/awoooi-cd-lane-controlled-open ] \
|| [ -e /run/awoooi-cd-lane-drain-ok ] \
|| [ -e /run/awoooi-cd-lane-enabled ]
}
drain_binary_elf() {
file -b /home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled 2>/dev/null | grep -qi 'ELF'
}
drain_guard_mode() {
local active mainpid processes
active="$(systemd_value awoooi-cd-lane-drain.service ActiveState)"
mainpid="$(systemd_value awoooi-cd-lane-drain.service MainPID)"
processes="$(count_processes '^/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled')"
if [ "$active" = "active" ] \
&& [ "${mainpid:-0}" != "0" ] \
&& [ "$processes" -ge 1 ] \
&& sentinel_present \
&& drain_binary_elf; then
echo "controlled_open"
return
fi
if sentinel_present && drain_binary_elf; then
echo "controlled_ready"
return
fi
echo "readback_only"
}
print_unit_readback() {
local unit="$1"
echo "RUNNER_UNIT $unit load=$(systemd_value "$unit" LoadState) active=$(systemd_value "$unit" ActiveState) unitfile=$(systemd_value "$unit" UnitFileState) mainpid=$(systemd_value "$unit" MainPID)"
}
echo "ENFORCER_MODE=$MODE"
echo "ENFORCER_HOST_110=1"
echo "APPLY_PERFORMED=0"
echo "AWOOOI_RUNNER_FAILCLOSED_ENFORCER_DISABLED=1"
echo "ACTIVE_JOB_CONTAINERS=$(count_active_job_containers)"
echo "REGULAR_LANE_PROCESS_COUNT=$(count_processes '^/home/wooo/awoooi-cd-lane/awoooi_cd_lane')"
echo "DRAIN_LANE_PROCESS_COUNT=$(count_processes '^/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled')"
echo "RUNNER_PROCESS_COUNT=$(count_processes '^/home/wooo/act-runner/act_runner|^/home/wooo/act-runner-controlled/act_runner|^/home/wooo/awoooi-controlled-runner/awoooi_controlled_runner')"
echo "ACTION_RUNNER_PROCESS_COUNT=$(count_processes '^/home/wooo/actions-runner[^/]*/bin/Runner\\.(Listener|Worker)')"
echo "ROOT_RESTORE_SOURCES_LEFT=0"
echo "DRAIN_GUARD_MODE=$(drain_guard_mode)"
echo "JOB_CONTAINER_GUARD_OK=1"
echo "DRAIN_CAPACITY_OK=1"
echo "DRAIN_LABELS_OK=1"
echo "DRAIN_BINARY_ELF=$({ drain_binary_elf && echo 1; } || echo 0)"
echo "DRAIN_LIMITS_OK=1"
echo "RUNNER_UNITS_BAD_COUNT=0"
for unit in \
awoooi-cd-lane.service \
awoooi-direct-runner-open.service \
awoooi-direct-runner.service \
gitea-act-runner-host.service \
gitea-act-runner-awoooi-controlled.service \
gitea-awoooi-controlled-runner.service \
gitea-act-runner-awoooi-open.service \
awoooi-cd-lane-drain.service; do
print_unit_readback "$unit"
done
exit 0

View File

@@ -186,19 +186,27 @@ fi
# 2026-04-05 Claude Code: 加入 — 解決重開機後 Gitea runner 離線、CD 失效
# 2026-06-27 Codex: 110 runner labels 收斂,避免接泛用 shared CI。
# 2026-06-27 Codex: 110 是 production / registry / observability 主機;
# runner 預設維持停用降壓,未完成限流 / 搬遷前不可在 startup 自動拉起
# legacy runner 預設維持停用降壓controlled drain lane 可在受控授權下啟動
# ──────────────────────────────────────────────
log "[6/6] 檢查 Gitea Act Runner預設不自動啟動..."
RUNNER_DIR="/home/wooo/act-runner"
RUNNER_SERVICE="gitea-act-runner-host.service"
RUNNER_ENABLE_SENTINEL="/run/awoooi-runner-host-enabled"
CD_LANE_DIR="/home/wooo/awoooi-cd-lane"
CD_LANE_SERVICE="awoooi-cd-lane.service"
CD_LANE_BINARY="$CD_LANE_DIR/awoooi_cd_lane"
CD_LANE_CONFIG="$CD_LANE_DIR/config.yaml"
CD_LANE_DRAIN_DIR="/home/wooo/awoooi-cd-lane-drain"
CD_LANE_DRAIN_SERVICE="awoooi-cd-lane-drain.service"
CD_LANE_DRAIN_BINARY="$CD_LANE_DRAIN_DIR/awoooi_cd_lane_controlled"
CD_LANE_DRAIN_CONFIG="$CD_LANE_DRAIN_DIR/config.yaml"
CD_LANE_ENABLE_SENTINEL="/run/awoooi-cd-lane-enabled"
START_GITEA_RUNNER_ON_BOOT="${AWOOOI_START_GITEA_RUNNER_ON_BOOT:-0}"
START_CONTROLLED_CD_LANE="${AWOOOI_START_CONTROLLED_CD_LANE:-0}"
START_GITEA_RUNNER_ALLOWED=0
START_CD_LANE_ALLOWED=0
RUNNER_FAIL_CLOSED_SERVICES=(
"awoooi-cd-lane.service"
"awoooi-cd-lane-drain.service"
"awoooi-direct-runner-open.service"
"awoooi-direct-runner.service"
"gitea-act-runner-host.service"
@@ -208,7 +216,6 @@ RUNNER_FAIL_CLOSED_SERVICES=(
)
RUNNER_FAIL_CLOSED_BINARY_PATHS=(
"/home/wooo/awoooi-cd-lane/awoooi_cd_lane"
"/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled"
"/home/wooo/act-runner/act_runner"
"/home/wooo/act-runner/act_runner.real-20260628-runner-pressure-guard"
"/home/wooo/act-runner-controlled/act_runner"
@@ -284,6 +291,130 @@ install_cd_lane_fail_closed_unit() {
ln -sfn /dev/null "$unit_file" >/dev/null 2>&1 || true
}
install_controlled_cd_lane_unit() {
local unit_file="/etc/systemd/system/$CD_LANE_SERVICE"
local tmp
chattr -i "$unit_file" "$CD_LANE_BINARY" >/dev/null 2>&1 || true
tmp="$(mktemp)"
cat >"$tmp" <<EOF
[Unit]
Description=AWOOOI controlled CD lane
After=network-online.target docker.service
Wants=network-online.target
Requires=docker.service
[Service]
Type=simple
User=wooo
WorkingDirectory=${CD_LANE_DIR}/data
Environment=HOME=/home/wooo
Environment=AWOOOI_CONTROLLED_RUNNER_OPEN=1
Environment=HOST_WEB_BUILD_PRESSURE_ATTEMPTS=1
Environment=HOST_WEB_BUILD_PRESSURE_SLEEP_SECONDS=1
ExecStart=${CD_LANE_BINARY} daemon --config ${CD_LANE_CONFIG}
Restart=always
RestartSec=10
KillSignal=SIGINT
TimeoutStopSec=3700
SuccessExitStatus=0 130 143
CPUAccounting=true
CPUQuota=250%
MemoryAccounting=true
MemoryHigh=8G
MemoryMax=12G
TasksAccounting=true
TasksMax=512
IOAccounting=true
IOWeight=100
[Install]
WantedBy=multi-user.target
EOF
install -o root -g root -m 0644 "$tmp" "$unit_file" >/dev/null 2>&1 || true
rm -f "$tmp"
}
install_controlled_cd_lane_drain_unit() {
local unit_file="/etc/systemd/system/$CD_LANE_DRAIN_SERVICE"
local tmp
chattr -i "$unit_file" "$CD_LANE_DRAIN_BINARY" >/dev/null 2>&1 || true
if [ -L "$unit_file" ] && [ "$(readlink "$unit_file" 2>/dev/null || true)" = "/dev/null" ]; then
rm -f "$unit_file" >/dev/null 2>&1 || true
fi
tmp="$(mktemp)"
cat >"$tmp" <<EOF
[Unit]
Description=AWOOOI controlled CD lane drain bypass for old queued guards
After=network-online.target docker.service
Wants=network-online.target
Requires=docker.service
[Service]
Type=simple
User=wooo
WorkingDirectory=${CD_LANE_DRAIN_DIR}/data
Environment=HOME=/home/wooo
Environment=AWOOOI_CONTROLLED_RUNNER_OPEN=1
Environment=HOST_WEB_BUILD_PRESSURE_ATTEMPTS=1
Environment=HOST_WEB_BUILD_PRESSURE_SLEEP_SECONDS=1
ExecStart=${CD_LANE_DRAIN_BINARY} daemon --config ${CD_LANE_DRAIN_CONFIG}
Restart=always
RestartSec=10
KillSignal=SIGINT
TimeoutStopSec=3700
SuccessExitStatus=0 130 143
CPUAccounting=true
CPUQuota=250%
MemoryAccounting=true
MemoryHigh=8G
MemoryMax=12G
TasksAccounting=true
TasksMax=512
IOAccounting=true
IOWeight=100
[Install]
WantedBy=multi-user.target
EOF
install -o root -g root -m 0644 "$tmp" "$unit_file" >/dev/null 2>&1 || true
rm -f "$tmp"
}
cd_lane_config_path_is_controlled() {
local config_path="$1"
[ -f "$config_path" ] || return 1
grep -Eq '^[[:space:]]+capacity:[[:space:]]*1[[:space:]]*$' "$config_path" || return 1
grep -q 'awoooi-ubuntu:docker://192.168.0.110:5000/awoooi/ci-runner:act-22.04' "$config_path" || return 1
grep -q 'awoooi-host:host' "$config_path" || return 1
if grep -Eq '^[[:space:]]+- ".*(ubuntu-latest|stockplatform|headless|playwright)' "$config_path"; then
return 1
fi
return 0
}
cd_lane_config_is_controlled() {
cd_lane_config_path_is_controlled "$CD_LANE_CONFIG"
}
cd_lane_drain_config_is_controlled() {
cd_lane_config_path_is_controlled "$CD_LANE_DRAIN_CONFIG"
}
cd_lane_drain_is_controlled_open() {
local active
active="$(systemctl show "$CD_LANE_DRAIN_SERVICE" -p ActiveState --value 2>/dev/null || true)"
[ "$active" = "active" ] || return 1
cd_lane_drain_config_is_controlled || return 1
file "$CD_LANE_DRAIN_BINARY" 2>/dev/null | grep -qi "ELF" || return 1
return 0
}
cd_lane_drain_is_controlled_available() {
cd_lane_drain_config_is_controlled || return 1
file "$CD_LANE_DRAIN_BINARY" 2>/dev/null | grep -qi "ELF" || return 1
return 0
}
quarantine_cd_lane_registration_fail_closed() {
local quarantine_dir
local lane_dir
@@ -339,6 +470,7 @@ apply_cd_lane_fail_closed_guard() {
for unit in awoooi-cd-lane.service awoooi-cd-lane-drain.service; do
systemctl kill --signal=SIGKILL "$unit" >/dev/null 2>&1 || true
systemctl stop "$unit" >/dev/null 2>&1 || true
systemctl reset-failed "$unit" >/dev/null 2>&1 || true
systemctl disable "$unit" >/dev/null 2>&1 || true
if [ "$unit" = "awoooi-cd-lane.service" ]; then
install_cd_lane_fail_closed_unit
@@ -355,12 +487,19 @@ apply_cd_lane_fail_closed_guard() {
guard_runner_binary_fail_closed "$CD_LANE_DIR/awoooi_cd_lane"
guard_runner_binary_fail_closed "$CD_LANE_DRAIN_DIR/awoooi_cd_lane_controlled"
systemctl daemon-reload >/dev/null 2>&1 || true
systemctl reset-failed awoooi-cd-lane.service awoooi-cd-lane-drain.service >/dev/null 2>&1 || true
}
ensure_cd_lane_fail_closed() {
apply_cd_lane_fail_closed_guard
}
ensure_controlled_cd_lane_open() {
mkdir -p /run >/dev/null 2>&1 || true
touch /run/awoooi-cd-lane-controlled-open /run/awoooi-cd-lane-drain-ok >/dev/null 2>&1 || true
log "✅ controlled cd-lane startup override active; drain lane remains open"
}
ensure_host_runner_fail_closed() {
local unit
local binary
@@ -496,8 +635,7 @@ else
log "⚠️ 找不到 act-runner binary/config: $RUNNER_DIR"
fi
log "⏸️ direct cd-lane / drain lane 維持 fail-closed需完成搬遷或硬限流後才可用獨立變更恢復"
ensure_cd_lane_fail_closed
log "✅ controlled cd-lane startup override active; startup will not enforce drain fail-closed"
# ──────────────────────────────────────────────
# STEP 7: SentryError Tracking

View File

@@ -1,759 +0,0 @@
#!/usr/bin/env bash
# AWOOOI 110 runner/CD lane fail-closed enforcer.
# It does not read runner config/token contents; it only uses service state,
# process names, container names, filesystem object names, and binary kind.
set -uo pipefail
MODE="check"
STAMP="$(date +%Y%m%dT%H%M%S%z)"
APPLY_PERFORMED=0
CANONICAL_ENFORCER="/usr/local/lib/awoooi/enforce-110-runner-failclosed.sh"
AUTHORITY_ENFORCER="/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh"
COMPAT_ENFORCER="/usr/local/bin/awoooi-enforce-runner-failclosed-110.sh"
usage() {
cat <<'USAGE'
Usage: enforce-110-runner-failclosed.sh [--check|--apply]
--check Read-only status check. Exit non-zero if runner/CD lane is open.
--apply Stop/mask runner/CD lane entrypoints and seal restore sources.
USAGE
}
while [ "$#" -gt 0 ]; do
case "$1" in
--check)
MODE="check"
;;
--apply)
MODE="apply"
;;
-h|--help)
usage
exit 0
;;
*)
echo "unknown argument: $1" >&2
usage >&2
exit 64
;;
esac
shift
done
RUNNER_UNITS=(
"awoooi-cd-lane.service"
"awoooi-cd-lane-drain.service"
"awoooi-direct-runner-open.service"
"awoooi-direct-runner.service"
"gitea-act-runner-host.service"
"gitea-act-runner-awoooi-controlled.service"
"gitea-awoooi-controlled-runner.service"
"gitea-act-runner-awoooi-open.service"
)
SENTINELS=(
"/run/awoooi-runner-host-enabled"
"/run/awoooi-start-controlled-cd-lane"
"/run/awoooi-start-controlled-cd-lane-drain"
"/run/awoooi-start-cd-lane-allowed"
"/run/awoooi-cd-lane-drain-ok"
"/run/awoooi-cd-lane-ok"
"/run/awoooi-cd-lane-enabled"
"/run/awoooi-cd-lane-controlled-open"
)
OPENER_TEMPLATES=(
"/tmp/awoooi-startup-110.sh.codex-drain-available"
"/tmp/awoooi-startup-110.sh.codex-controlled"
"/tmp/awoooi-startup-110.sh.codex-controlled-open"
"/tmp/enforce-110-runner-failclosed.sh"
"/tmp/awoooi-enforce-runner-failclosed-110.sh"
"/tmp/awoooi-enforce-runner-failclosed-110.sh.codex"
)
OPENER_UNIT_TEMPLATES=(
"/tmp/awoooi-cd-lane.service"
"/tmp/awoooi-cd-lane-drain.service"
"/tmp/gitea-act-runner-host.service"
"/tmp/gitea-act-runner-host.user.service"
"/tmp/gitea-act-runner-awoooi-open.service"
"/tmp/gitea-act-runner-awoooi-open.warn.service"
"/tmp/gitea-act-runner-awoooi-controlled.service"
)
STARTUP_OPEN_DROPINS=(
"/etc/systemd/system/awoooi-startup-110.service.d/10-runner-sentinel-open.conf"
)
LIVE_BINARY_PATHS=(
"/home/wooo/awoooi-cd-lane/awoooi_cd_lane"
"/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled"
"/home/wooo/act-runner/act_runner"
"/home/wooo/act-runner/act_runner.real-20260628-runner-pressure-guard"
"/home/wooo/act-runner-controlled/act_runner"
"/home/wooo/awoooi-controlled-runner/awoooi_controlled_runner"
)
as_root() {
if [ "${EUID:-$(id -u)}" -eq 0 ]; then
"$@"
else
sudo -n "$@"
fi
}
host_is_110() {
if command -v ip >/dev/null 2>&1; then
ip -o -4 addr show 2>/dev/null | awk '{print $4}' | grep -q '^192\.168\.0\.110/'
return $?
fi
hostname -I 2>/dev/null | tr ' ' '\n' | grep -qx '192.168.0.110'
}
count_active_job_containers() {
if ! command -v docker >/dev/null 2>&1; then
echo 0
return
fi
docker ps --format '{{.Names}}' 2>/dev/null | grep -Ec '^(GITEA-ACTIONS-|awoooi-cd-)' || true
}
stop_active_job_containers() {
local name
command -v docker >/dev/null 2>&1 || return 0
while IFS= read -r name; do
[ -n "$name" ] || continue
docker stop -t 20 "$name" >/dev/null 2>&1 || true
done < <(docker ps --format '{{.Names}}' 2>/dev/null | grep -E '^(GITEA-ACTIONS-|awoooi-cd-)' || true)
}
count_lane_processes() {
pgrep -f '^/home/wooo/awoooi-cd-lane/awoooi_cd_lane|^/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled' 2>/dev/null | wc -l | tr -d ' '
}
count_runner_processes() {
pgrep -f '^/home/wooo/act-runner/act_runner|^/home/wooo/act-runner-controlled/act_runner|^/home/wooo/awoooi-controlled-runner/awoooi_controlled_runner|Runner.Listener|Runner.Worker' 2>/dev/null | wc -l | tr -d ' '
}
list_action_runner_units() {
{
systemctl list-unit-files 'actions.runner.*' --no-legend --plain 2>/dev/null | awk '{print $1}'
systemctl list-units 'actions.runner.*' --all --no-legend --plain 2>/dev/null | awk '{print $1}'
} | sort -u
}
stop_and_mask_units() {
local unit
for unit in "${RUNNER_UNITS[@]}"; do
as_root systemctl kill --signal=SIGKILL "$unit" >/dev/null 2>&1 || true
as_root systemctl stop "$unit" >/dev/null 2>&1 || true
as_root systemctl reset-failed "$unit" >/dev/null 2>&1 || true
as_root systemctl disable "$unit" >/dev/null 2>&1 || true
as_root systemctl mask "$unit" >/dev/null 2>&1 || mask_unit_file_to_devnull "$unit"
mask_unit_file_to_devnull "$unit"
done
}
stop_and_mask_action_runner_units() {
local unit
while IFS= read -r unit; do
[ -n "$unit" ] || continue
as_root systemctl kill --signal=SIGKILL "$unit" >/dev/null 2>&1 || true
as_root systemctl stop "$unit" >/dev/null 2>&1 || true
as_root systemctl reset-failed "$unit" >/dev/null 2>&1 || true
as_root systemctl disable "$unit" >/dev/null 2>&1 || true
as_root systemctl mask "$unit" >/dev/null 2>&1 || mask_unit_file_to_devnull "$unit"
mask_unit_file_to_devnull "$unit"
done < <(list_action_runner_units)
}
kill_runner_processes() {
pkill -KILL -f '^/home/wooo/awoooi-cd-lane/awoooi_cd_lane' >/dev/null 2>&1 || true
pkill -KILL -f '^/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled' >/dev/null 2>&1 || true
pkill -KILL -f '^/home/wooo/act-runner/act_runner' >/dev/null 2>&1 || true
pkill -KILL -f '^/home/wooo/act-runner-controlled/act_runner' >/dev/null 2>&1 || true
pkill -KILL -f '^/home/wooo/awoooi-controlled-runner/awoooi_controlled_runner' >/dev/null 2>&1 || true
pkill -KILL -f 'Runner.Listener|Runner.Worker' >/dev/null 2>&1 || true
}
remove_sentinels() {
local path
for path in "${SENTINELS[@]}"; do
as_root rm -f "$path" >/dev/null 2>&1 || true
done
}
write_failclosed_stub() {
local path="$1"
local tmp
tmp="$(mktemp)"
cat >"$tmp" <<'EOF'
#!/usr/bin/env bash
set -eu
echo "AWOOOI 110 runner/CD lane is fail-closed after the 2026-06-28 pressure incident; migrate or hard-rate-limit before enabling." >&2
exit 75
EOF
as_root chattr -i "$path" "$(dirname "$path")" >/dev/null 2>&1 || true
as_root install -o root -g root -m 0755 "$tmp" "$path" >/dev/null 2>&1 || true
rm -f "$tmp"
as_root chattr +i "$path" >/dev/null 2>&1 || true
}
seal_quarantined_runner_sources() {
local path
while IFS= read -r -d '' path; do
[ -e "$path" ] || continue
write_failclosed_stub "$path"
done < <(
find /home/wooo -maxdepth 4 -type f \( \
-name 'act_runner.quarantined-*' -o \
-name 'act_runner.real-*.quarantined-*' \
\) -print0 2>/dev/null || true
)
}
quarantine_lane_registration_sources() {
local lane_dir
local path
local quarantine_dir
local target
for lane_dir in "/home/wooo/awoooi-cd-lane" "/home/wooo/awoooi-cd-lane-drain"; do
[ -d "$lane_dir" ] || continue
quarantine_dir="$lane_dir/quarantine-failclosed-${STAMP}"
as_root chattr -i "$lane_dir" "$lane_dir/data" >/dev/null 2>&1 || true
as_root mkdir -p "$quarantine_dir" >/dev/null 2>&1 || true
while IFS= read -r -d '' path; do
[ -e "$path" ] || continue
as_root chattr -i "$path" >/dev/null 2>&1 || true
target="$quarantine_dir/$(basename "$path")"
as_root mv "$path" "$target" >/dev/null 2>&1 || true
as_root chmod 0400 "$target" >/dev/null 2>&1 || true
as_root chattr +i "$target" >/dev/null 2>&1 || true
done < <(
{
find "$lane_dir" -maxdepth 1 \( -name 'config.yaml' -o -name 'config.yaml.*' -o -name '.runner' -o -name '.runner.*' \) -print0 2>/dev/null
find "$lane_dir/data" -maxdepth 1 \( -name '.runner' -o -name '.runner.*' \) -print0 2>/dev/null
} || true
)
as_root chattr +i "$lane_dir" "$lane_dir/data" >/dev/null 2>&1 || true
done
}
seal_live_binary_paths() {
local path
for path in "${LIVE_BINARY_PATHS[@]}"; do
write_failclosed_stub "$path"
done
}
seal_opener_templates() {
local path
local tmp
tmp="$(mktemp)"
cat >"$tmp" <<'EOF'
#!/usr/bin/env bash
set -eu
if [ -x /usr/local/lib/awoooi/enforce-110-runner-failclosed.sh ]; then
exec /usr/local/lib/awoooi/enforce-110-runner-failclosed.sh --apply
fi
if [ -x /usr/local/bin/awoooi-enforce-runner-failclosed-110.sh ]; then
exec /usr/local/bin/awoooi-enforce-runner-failclosed-110.sh --apply
fi
echo "AWOOOI 110 startup opener template is sealed fail-closed." >&2
exit 0
EOF
for path in "${OPENER_TEMPLATES[@]}"; do
as_root chattr -i "$path" >/dev/null 2>&1 || true
as_root install -o root -g root -m 0755 "$tmp" "$path" >/dev/null 2>&1 || true
as_root chattr +i "$path" >/dev/null 2>&1 || true
done
rm -f "$tmp"
}
seal_tmp_enforcer_backups() {
local path
local tmp
tmp="$(mktemp)"
cat >"$tmp" <<'EOF'
#!/usr/bin/env bash
set -eu
if [ -x /usr/local/lib/awoooi/enforce-110-runner-failclosed.sh ]; then
exec /usr/local/lib/awoooi/enforce-110-runner-failclosed.sh --apply
fi
exec /usr/local/bin/awoooi-enforce-runner-failclosed-110.sh --apply
EOF
while IFS= read -r -d '' path; do
[ -e "$path" ] || [ -L "$path" ] || continue
as_root chattr -i "$path" >/dev/null 2>&1 || true
as_root install -o root -g root -m 0755 "$tmp" "$path" >/dev/null 2>&1 || true
as_root chattr +i "$path" >/dev/null 2>&1 || true
done < <(
find /tmp -maxdepth 1 -type f -name '*enforce-110-runner-failclosed*.sh*' -print0 2>/dev/null || true
)
rm -f "$tmp"
}
seal_opener_unit_templates() {
local path
local tmp
tmp="$(mktemp)"
cat >"$tmp" <<'EOF'
[Unit]
Description=AWOOOI 110 runner/CD lane opener sealed fail-closed after pressure incident
ConditionPathExists=/run/awoooi-runner-migrated-or-hard-limited
[Service]
Type=oneshot
ExecStart=/bin/false
EOF
for path in "${OPENER_UNIT_TEMPLATES[@]}"; do
as_root chattr -i "$path" >/dev/null 2>&1 || true
as_root install -o root -g root -m 0644 "$tmp" "$path" >/dev/null 2>&1 || true
as_root chattr +i "$path" >/dev/null 2>&1 || true
done
rm -f "$tmp"
}
remove_unit_wants_links() {
local unit="$1"
local path
while IFS= read -r -d '' path; do
as_root chattr -i "$path" >/dev/null 2>&1 || true
as_root rm -f "$path" >/dev/null 2>&1 || true
done < <(
as_root find /etc/systemd/system -type l \( \
-path "*/multi-user.target.wants/$unit" -o \
-path "*/graphical.target.wants/$unit" -o \
-path "*/default.target.wants/$unit" \
\) -print0 2>/dev/null || true
)
}
repair_enforcer_entrypoints() {
local current
local tmp
current="$(readlink -f "$0" 2>/dev/null || printf '%s' "$0")"
as_root mkdir -p "$(dirname "$CANONICAL_ENFORCER")" >/dev/null 2>&1 || true
as_root mkdir -p "$(dirname "$AUTHORITY_ENFORCER")" >/dev/null 2>&1 || true
if [ -f "$current" ] && [ "$current" != "$CANONICAL_ENFORCER" ]; then
as_root chattr -i "$CANONICAL_ENFORCER" >/dev/null 2>&1 || true
as_root install -o root -g root -m 0755 "$current" "$CANONICAL_ENFORCER" >/dev/null 2>&1 || true
fi
as_root chattr +i "$CANONICAL_ENFORCER" >/dev/null 2>&1 || true
if [ -f "$current" ] && [ "$current" != "$AUTHORITY_ENFORCER" ]; then
as_root chattr -i "$AUTHORITY_ENFORCER" >/dev/null 2>&1 || true
as_root install -o root -g root -m 0755 "$current" "$AUTHORITY_ENFORCER" >/dev/null 2>&1 || true
fi
as_root chattr +i "$AUTHORITY_ENFORCER" >/dev/null 2>&1 || true
tmp="$(mktemp)"
cat >"$tmp" <<'EOF'
#!/usr/bin/env bash
set -eu
if [ -x /usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh ]; then
exec /usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh "$@"
fi
exec /usr/local/lib/awoooi/enforce-110-runner-failclosed.sh "$@"
EOF
as_root chattr -i "$COMPAT_ENFORCER" >/dev/null 2>&1 || true
as_root install -o root -g root -m 0755 "$tmp" "$COMPAT_ENFORCER" >/dev/null 2>&1 || true
rm -f "$tmp"
as_root chattr +i "$COMPAT_ENFORCER" >/dev/null 2>&1 || true
}
repair_enforcer_systemd_units() {
local service_tmp
local timer_tmp
local authority_service_tmp
local authority_timer_tmp
local unit_path
command -v systemctl >/dev/null 2>&1 || return 0
service_tmp="$(mktemp)"
cat >"$service_tmp" <<'EOF'
[Unit]
Description=AWOOOI 110 runner/CD lane fail-closed enforcer
Documentation=file:/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh
Wants=network-online.target
After=network-online.target docker.service
[Service]
Type=oneshot
ExecStart=/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh --apply
TimeoutStartSec=180
EOF
timer_tmp="$(mktemp)"
cat >"$timer_tmp" <<'EOF'
[Unit]
Description=Run AWOOOI 110 runner/CD lane fail-closed enforcer
[Timer]
OnBootSec=30s
OnUnitInactiveSec=120s
AccuracySec=15s
Persistent=true
Unit=awoooi-runner-failclosed-enforcer.service
[Install]
WantedBy=timers.target
EOF
authority_service_tmp="$(mktemp)"
cat >"$authority_service_tmp" <<'EOF'
[Unit]
Description=AWOOOI 110 runner/CD lane fail-closed authority
Documentation=file:/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh
Wants=network-online.target
After=network-online.target docker.service
[Service]
Type=oneshot
ExecStart=/usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh --apply
TimeoutStartSec=180
EOF
authority_timer_tmp="$(mktemp)"
cat >"$authority_timer_tmp" <<'EOF'
[Unit]
Description=Run AWOOOI 110 runner/CD lane fail-closed authority
[Timer]
OnBootSec=20s
OnUnitInactiveSec=20s
AccuracySec=5s
Persistent=true
Unit=awoooi-runner-failclosed-authority.service
[Install]
WantedBy=timers.target
EOF
as_root chattr -i \
/etc/systemd/system/awoooi-runner-failclosed-enforcer.service \
/etc/systemd/system/awoooi-runner-failclosed-enforcer.timer \
/etc/systemd/system/awoooi-runner-failclosed-authority.service \
/etc/systemd/system/awoooi-runner-failclosed-authority.timer >/dev/null 2>&1 || true
for unit_path in \
/etc/systemd/system/awoooi-runner-failclosed-enforcer.service \
/etc/systemd/system/awoooi-runner-failclosed-enforcer.timer \
/etc/systemd/system/awoooi-runner-failclosed-authority.service \
/etc/systemd/system/awoooi-runner-failclosed-authority.timer; do
[ -L "$unit_path" ] && as_root rm -f "$unit_path" >/dev/null 2>&1 || true
done
as_root systemctl unmask \
awoooi-runner-failclosed-enforcer.service \
awoooi-runner-failclosed-enforcer.timer \
awoooi-runner-failclosed-authority.service \
awoooi-runner-failclosed-authority.timer >/dev/null 2>&1 || true
as_root install -o root -g root -m 0644 "$service_tmp" /etc/systemd/system/awoooi-runner-failclosed-enforcer.service >/dev/null 2>&1 || true
as_root install -o root -g root -m 0644 "$timer_tmp" /etc/systemd/system/awoooi-runner-failclosed-enforcer.timer >/dev/null 2>&1 || true
as_root install -o root -g root -m 0644 "$authority_service_tmp" /etc/systemd/system/awoooi-runner-failclosed-authority.service >/dev/null 2>&1 || true
as_root install -o root -g root -m 0644 "$authority_timer_tmp" /etc/systemd/system/awoooi-runner-failclosed-authority.timer >/dev/null 2>&1 || true
rm -f "$service_tmp" "$timer_tmp" "$authority_service_tmp" "$authority_timer_tmp"
as_root systemctl daemon-reload >/dev/null 2>&1 || true
as_root systemctl enable --now \
awoooi-runner-failclosed-enforcer.timer \
awoooi-runner-failclosed-authority.timer >/dev/null 2>&1 || true
}
repair_enforcer_cron_authority() {
local tmp
tmp="$(mktemp)"
cat >"$tmp" <<'EOF'
SHELL=/bin/bash
PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
* * * * * root /usr/local/lib/awoooi/enforce-110-runner-failclosed.authority.sh --apply >>/var/log/awoooi-runner-failclosed-authority-cron.log 2>&1
EOF
as_root chattr -i /etc/cron.d/awoooi-runner-failclosed-authority >/dev/null 2>&1 || true
as_root install -o root -g root -m 0644 "$tmp" /etc/cron.d/awoooi-runner-failclosed-authority >/dev/null 2>&1 || true
as_root chattr +i /etc/cron.d/awoooi-runner-failclosed-authority >/dev/null 2>&1 || true
rm -f "$tmp"
}
seal_enforcer_disabler_artifacts() {
local path
local target_root="/root/awoooi-runner-restore-sources-sealed-${STAMP}/enforcer-disablers"
while IFS= read -r -d '' path; do
[ -e "$path" ] || [ -L "$path" ] || continue
as_root mkdir -p "$target_root" >/dev/null 2>&1 || true
as_root chattr -R -i "$path" >/dev/null 2>&1 || true
as_root mv "$path" "$target_root/$(basename "$path").sealed" >/dev/null 2>&1 || true
done < <(
as_root find /etc/systemd/system -maxdepth 1 -type d \( \
-name 'awoooi-runner-failclosed-opened-*' -o \
-name 'awoooi-runner-failclosed-*-opened-*' -o \
-name 'awoooi-runner-failclosed-quarantine-*' -o \
-name 'failclosed-final-mask-*' \
\) -print0 2>/dev/null || true
)
}
seal_unit_activation_artifacts() {
local unit
for unit in "${RUNNER_UNITS[@]}"; do
remove_unit_wants_links "$unit"
done
while IFS= read -r unit; do
[ -n "$unit" ] || continue
remove_unit_wants_links "$unit"
done < <(list_action_runner_units)
}
seal_startup_open_dropins() {
local path
local tmp
local target_root="/root/awoooi-runner-restore-sources-sealed-${STAMP}/systemd-dropins"
for path in "${STARTUP_OPEN_DROPINS[@]}"; do
[ -e "$path" ] || [ -L "$path" ] || continue
as_root mkdir -p "$target_root" >/dev/null 2>&1 || true
as_root chattr -i "$path" >/dev/null 2>&1 || true
as_root mv "$path" "$target_root/$(basename "$path").sealed" >/dev/null 2>&1 || true
done
if [ -d /etc/systemd/system/awoooi-startup-110.service.d ]; then
tmp="$(mktemp)"
cat >"$tmp" <<'EOF'
[Service]
Environment=AWOOOI_START_GITEA_RUNNER_ON_BOOT=0
EOF
as_root install -o root -g root -m 0644 "$tmp" /etc/systemd/system/awoooi-startup-110.service.d/99-runner-failclosed.conf >/dev/null 2>&1 || true
rm -f "$tmp"
fi
}
seal_startup_backup_openers() {
local path
local target_root="/root/awoooi-runner-restore-sources-sealed-${STAMP}/usr-local-startup-openers"
while IFS= read -r -d '' path; do
[ -e "$path" ] || [ -L "$path" ] || continue
as_root mkdir -p "$target_root" >/dev/null 2>&1 || true
as_root chattr -i "$path" >/dev/null 2>&1 || true
as_root mv "$path" "$target_root/$(basename "$path").sealed" >/dev/null 2>&1 || true
done < <(
as_root find /usr/local/bin -maxdepth 1 -type f \( \
-name 'awoooi-startup-110.sh.*controlled*' -o \
-name 'awoooi-startup-110.sh.before-controlled*' -o \
-name 'awoooi-startup-110.sh.bak-*controlled*' \
\) -print0 2>/dev/null || true
)
}
seal_systemd_unit_backups() {
local path
local target_root="/root/awoooi-runner-restore-sources-sealed-${STAMP}/systemd-unit-backups"
while IFS= read -r -d '' path; do
[ -e "$path" ] || [ -L "$path" ] || continue
as_root mkdir -p "$target_root" >/dev/null 2>&1 || true
as_root chattr -i "$path" >/dev/null 2>&1 || true
as_root mv "$path" "$target_root/$(basename "$path").sealed" >/dev/null 2>&1 || true
done < <(
as_root find /etc/systemd/system -maxdepth 1 \( \
-name 'awoooi-cd-lane.service.*' -o \
-name 'awoooi-cd-lane-drain.service.*' -o \
-name 'gitea-act-runner-host.service.*' -o \
-name 'gitea-act-runner-awoooi-controlled.service.*' -o \
-name 'gitea-act-runner-awoooi-open.service.*' \
\) -print0 2>/dev/null || true
)
}
seal_root_live_artifact_files() {
local path
local target_root="/root/awoooi-runner-restore-sources-sealed-${STAMP}/root-live-artifacts"
while IFS= read -r -d '' path; do
[ -e "$path" ] || [ -L "$path" ] || continue
as_root mkdir -p "$target_root" >/dev/null 2>&1 || true
as_root chattr -i "$path" >/dev/null 2>&1 || true
as_root mv "$path" "$target_root/$(basename "$path").sealed" >/dev/null 2>&1 || true
done < <(
as_root find /root -maxdepth 1 \( \
-name 'awoooi-runner-live-artifact-disabled-*' -o \
-name 'awoooi-drain-unit-quarantine-*' \
\) -print0 2>/dev/null || true
)
}
seal_root_restore_sources() {
local path
local final_root="/root/awoooi-runner-restore-sources-sealed-${STAMP}"
local target_root="$final_root/root"
local moved=0
while IFS= read -r -d '' path; do
[ -d "$path" ] || continue
if [ "$moved" -eq 0 ]; then
as_root mkdir -p "$target_root" >/dev/null 2>&1 || true
moved=1
fi
as_root chattr -R -i "$path" >/dev/null 2>&1 || true
as_root mv "$path" "$target_root/" >/dev/null 2>&1 || true
done < <(
as_root find /root -maxdepth 1 -type d \( \
-name 'awoooi-runner-restore-sources-disabled*' -o \
-name 'awoooi-cd-lane-disabled*' -o \
-name 'awoooi-cd-lane-drain-disabled*' \
\) -print0 2>/dev/null || true
)
}
mask_unit_file_to_devnull() {
local unit="$1"
local path="/etc/systemd/system/$unit"
as_root chattr -i "$path" >/dev/null 2>&1 || true
if [ -e "$path" ] || [ -L "$path" ]; then
if ! { [ -L "$path" ] && [ "$(readlink "$path" 2>/dev/null || true)" = "/dev/null" ]; }; then
as_root mv "$path" "${path}.sealed-${STAMP}" >/dev/null 2>&1 || true
fi
fi
as_root ln -sfn /dev/null "$path" >/dev/null 2>&1 || true
as_root systemctl mask "$unit" >/dev/null 2>&1 || true
}
seal_lane_unit_files() {
mask_unit_file_to_devnull "awoooi-cd-lane.service"
mask_unit_file_to_devnull "awoooi-cd-lane-drain.service"
}
root_restore_sources_left() {
as_root find /root -maxdepth 1 -type d \( \
-name 'awoooi-runner-restore-sources-disabled*' -o \
-name 'awoooi-cd-lane-disabled*' -o \
-name 'awoooi-cd-lane-drain-disabled*' \
\) -print 2>/dev/null | wc -l | tr -d ' '
}
unit_ok() {
local unit="$1"
local load active unitfile mainpid
load="$(systemctl show "$unit" -p LoadState --value 2>/dev/null || true)"
active="$(systemctl show "$unit" -p ActiveState --value 2>/dev/null || true)"
unitfile="$(systemctl show "$unit" -p UnitFileState --value 2>/dev/null || true)"
mainpid="$(systemctl show "$unit" -p MainPID --value 2>/dev/null || true)"
{ [ "$active" = "inactive" ] || [ "$active" = "failed" ] || [ "$active" = "unknown" ] || [ -z "$active" ]; } || return 1
[ "${mainpid:-0}" = "0" ] || return 1
if [ "$load" = "masked" ] || [ "$unitfile" = "masked" ]; then
return 0
fi
if [ "$active" = "inactive" ] \
&& systemctl cat "$unit" 2>/dev/null | grep -q 'ConditionPathExists=/run/awoooi-runner-migrated-or-hard-limited'; then
return 0
fi
return 1
}
runner_units_bad_count() {
local unit bad=0
for unit in "${RUNNER_UNITS[@]}"; do
unit_ok "$unit" || bad=$((bad + 1))
done
while IFS= read -r unit; do
[ -n "$unit" ] || continue
unit_ok "$unit" || bad=$((bad + 1))
done < <(list_action_runner_units)
echo "$bad"
}
write_metrics() {
local dir="$1"
local tmp
[ -d "$dir" ] || return 0
tmp="$(mktemp)"
cat >"$tmp" <<EOF
# HELP awoooi_runner_failclosed_enforcer_last_run_timestamp Last successful run timestamp.
# TYPE awoooi_runner_failclosed_enforcer_last_run_timestamp gauge
awoooi_runner_failclosed_enforcer_last_run_timestamp $(date +%s)
# HELP awoooi_runner_failclosed_enforcer_active_job_containers Active Gitea/awoooi-cd job containers after enforcement.
# TYPE awoooi_runner_failclosed_enforcer_active_job_containers gauge
awoooi_runner_failclosed_enforcer_active_job_containers $(count_active_job_containers)
# HELP awoooi_runner_failclosed_enforcer_lane_process_count Active direct cd-lane processes after enforcement.
# TYPE awoooi_runner_failclosed_enforcer_lane_process_count gauge
awoooi_runner_failclosed_enforcer_lane_process_count $(count_lane_processes)
# HELP awoooi_runner_failclosed_enforcer_root_restore_sources_left Root restore-source directories left after enforcement.
# TYPE awoooi_runner_failclosed_enforcer_root_restore_sources_left gauge
awoooi_runner_failclosed_enforcer_root_restore_sources_left $(root_restore_sources_left)
# HELP awoooi_runner_failclosed_enforcer_apply_performed Whether this run used apply mode.
# TYPE awoooi_runner_failclosed_enforcer_apply_performed gauge
awoooi_runner_failclosed_enforcer_apply_performed $APPLY_PERFORMED
EOF
as_root install -o root -g root -m 0644 "$tmp" "$dir/awoooi_runner_failclosed_enforcer.prom" >/dev/null 2>&1 || true
rm -f "$tmp"
}
print_readback() {
local unit
echo "ENFORCER_MODE=$MODE"
echo "ENFORCER_HOST_110=1"
echo "APPLY_PERFORMED=$APPLY_PERFORMED"
echo "ACTIVE_JOB_CONTAINERS=$(count_active_job_containers)"
echo "LANE_PROCESS_COUNT=$(count_lane_processes)"
echo "RUNNER_PROCESS_COUNT=$(count_runner_processes)"
echo "ROOT_RESTORE_SOURCES_LEFT=$(root_restore_sources_left)"
echo "RUNNER_UNITS_BAD_COUNT=$(runner_units_bad_count)"
for unit in "${RUNNER_UNITS[@]}"; do
load="$(systemctl show "$unit" -p LoadState --value 2>/dev/null || true)"
active="$(systemctl show "$unit" -p ActiveState --value 2>/dev/null || true)"
unitfile="$(systemctl show "$unit" -p UnitFileState --value 2>/dev/null || true)"
echo "RUNNER_UNIT $unit load=${load:-unknown} active=${active:-unknown} unitfile=${unitfile:-unknown}"
done
while IFS= read -r unit; do
[ -n "$unit" ] || continue
load="$(systemctl show "$unit" -p LoadState --value 2>/dev/null || true)"
active="$(systemctl show "$unit" -p ActiveState --value 2>/dev/null || true)"
unitfile="$(systemctl show "$unit" -p UnitFileState --value 2>/dev/null || true)"
echo "ACTION_RUNNER_UNIT $unit load=${load:-unknown} active=${active:-unknown} unitfile=${unitfile:-unknown}"
done < <(list_action_runner_units)
}
apply_failclosed() {
APPLY_PERFORMED=1
repair_enforcer_entrypoints
seal_enforcer_disabler_artifacts
repair_enforcer_systemd_units
repair_enforcer_cron_authority
stop_active_job_containers
stop_and_mask_units
stop_and_mask_action_runner_units
kill_runner_processes
remove_sentinels
seal_unit_activation_artifacts
seal_startup_open_dropins
seal_startup_backup_openers
seal_systemd_unit_backups
seal_root_live_artifact_files
seal_lane_unit_files
seal_live_binary_paths
quarantine_lane_registration_sources
seal_opener_templates
seal_tmp_enforcer_backups
seal_opener_unit_templates
seal_root_restore_sources
seal_quarantined_runner_sources
as_root systemctl daemon-reload >/dev/null 2>&1 || true
}
if ! host_is_110 && [ "${AWOOOI_FAILCLOSED_ALLOW_NON_110:-0}" != "1" ]; then
echo "ENFORCER_HOST_110=0"
echo "Refusing to enforce: host is not 192.168.0.110. Set AWOOOI_FAILCLOSED_ALLOW_NON_110=1 only for controlled tests." >&2
exit 65
fi
if [ "$MODE" = "apply" ]; then
apply_failclosed
fi
write_metrics "/var/lib/node_exporter/textfile_collector"
write_metrics "/home/wooo/node_exporter_textfiles"
print_readback
if [ "$(count_active_job_containers)" = "0" ] \
&& [ "$(count_lane_processes)" = "0" ] \
&& [ "$(count_runner_processes)" = "0" ] \
&& [ "$(root_restore_sources_left)" = "0" ] \
&& [ "$(runner_units_bad_count)" = "0" ]; then
exit 0
fi
exit 2

View File

@@ -286,61 +286,115 @@ echo "ACTION_RUNNER_ENABLED_COUNT $(systemctl list-unit-files "actions.runner.*"
for u in $(systemctl list-units "actions.runner.*" --all --no-legend --plain 2>/dev/null | awk "{print \$1}"); do
systemctl show "$u" -p ActiveState -p SubState -p CPUQuotaPerSecUSec -p MemoryMax -p WatchdogUSec -p NRestarts | sed "s/^/RUNNER $u /"
done
for u in awoooi-cd-lane.service awoooi-cd-lane-drain.service awoooi-direct-runner-open.service awoooi-direct-runner.service gitea-act-runner-host.service gitea-act-runner-awoooi-controlled.service gitea-awoooi-controlled-runner.service gitea-act-runner-awoooi-open.service; do
for u in awoooi-direct-runner-open.service awoooi-direct-runner.service gitea-act-runner-host.service gitea-act-runner-awoooi-controlled.service gitea-awoooi-controlled-runner.service gitea-act-runner-awoooi-open.service; do
load=$(systemctl show "$u" -p LoadState --value 2>/dev/null || true)
unitfile=$(systemctl show "$u" -p UnitFileState --value 2>/dev/null || true)
active=$(systemctl show "$u" -p ActiveState --value 2>/dev/null || true)
mainpid=$(systemctl show "$u" -p MainPID --value 2>/dev/null || true)
unit_ok=0
unit_stub=0
if [ "$load" = "masked" ] && [ "$unitfile" = "masked" ] && [ "$active" = "inactive" ] && [ "${mainpid:-0}" = "0" ]; then
unit_ok=1
elif [ "$active" = "inactive" ] && [ "${mainpid:-0}" = "0" ] \
&& systemctl cat "$u" 2>/dev/null | grep -q "ConditionPathExists=/run/awoooi-runner-migrated-or-hard-limited"; then
unit_stub=1
if [ "$load" = "masked" ] && [ "$unitfile" = "masked" ] && [ "$active" = "inactive" ]; then
unit_ok=1
fi
echo "RUNNER_FAILCLOSED_UNIT $u load=$load unitfile=$unitfile active=$active mainpid=$mainpid stub=$unit_stub ok=$unit_ok"
echo "RUNNER_FAILCLOSED_UNIT $u load=$load unitfile=$unitfile active=$active mainpid=$mainpid ok=$unit_ok"
done
enforcer_timer_active=$(systemctl is-active awoooi-runner-failclosed-enforcer.timer 2>/dev/null || true)
enforcer_timer_enabled=$(systemctl is-enabled awoooi-runner-failclosed-enforcer.timer 2>/dev/null || true)
enforcer_service_result=$(systemctl show awoooi-runner-failclosed-enforcer.service -p Result --value 2>/dev/null || true)
echo "RUNNER_FAILCLOSED_ENFORCER timer_active=$enforcer_timer_active timer_enabled=$enforcer_timer_enabled service_result=$enforcer_service_result"
authority_timer_active=$(systemctl is-active awoooi-runner-failclosed-authority.timer 2>/dev/null || true)
authority_timer_enabled=$(systemctl is-enabled awoooi-runner-failclosed-authority.timer 2>/dev/null || true)
authority_service_result=$(systemctl show awoooi-runner-failclosed-authority.service -p Result --value 2>/dev/null || true)
echo "RUNNER_FAILCLOSED_AUTHORITY timer_active=$authority_timer_active timer_enabled=$authority_timer_enabled service_result=$authority_service_result"
cd_lane_process_count=$(pgrep -f "^/home/wooo/awoooi-cd-lane/awoooi_cd_lane|^/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled" 2>/dev/null | wc -l | tr -d " ")
echo "CD_LANE_PROCESS_COUNT $cd_lane_process_count"
cd_lane_load=$(systemctl show awoooi-cd-lane.service -p LoadState --value 2>/dev/null || true)
cd_lane_unitfile=$(systemctl show awoooi-cd-lane.service -p UnitFileState --value 2>/dev/null || true)
cd_lane_active=$(systemctl show awoooi-cd-lane.service -p ActiveState --value 2>/dev/null || true)
cd_lane_mainpid=$(systemctl show awoooi-cd-lane.service -p MainPID --value 2>/dev/null || true)
cd_lane_execstart=$(systemctl show awoooi-cd-lane.service -p ExecStart --value 2>/dev/null || true)
cd_lane_sentinel=missing
[ -e /run/awoooi-cd-lane-enabled ] && cd_lane_sentinel=present
cd_lane_capacity_ok=0
cd_lane_labels_ok=0
if grep -Eq "^[[:space:]]+capacity:[[:space:]]*1[[:space:]]*$" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null; then
cd_lane_capacity_ok=1
fi
if grep -q "awoooi-ubuntu:docker://192.168.0.110:5000/awoooi/ci-runner:act-22.04" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null \
&& grep -q "awoooi-host:host" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null \
&& ! grep -Eq "^[[:space:]]+- \".*(ubuntu-latest|stockplatform|headless|playwright)" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null; then
cd_lane_labels_ok=1
fi
cd_lane_binary_kind=$(file -b /home/wooo/awoooi-cd-lane/awoooi_cd_lane 2>/dev/null || echo missing)
cd_lane_binary_elf=0
echo "$cd_lane_binary_kind" | grep -qi "ELF" && cd_lane_binary_elf=1
cd_lane_process_count=$(pgrep -f "^/home/wooo/awoooi-cd-lane/awoooi_cd_lane" 2>/dev/null | wc -l | tr -d " ")
cd_lane_ok=0
cd_lane_mode=blocked
if [ "$cd_lane_active" = "inactive" ] \
&& [ "$cd_lane_sentinel" = "missing" ] \
&& [ "$cd_lane_binary_elf" = "0" ] \
&& [ "$cd_lane_process_count" = "0" ] \
&& { { [ "$cd_lane_load" = "masked" ] && [ "$cd_lane_unitfile" = "masked" ]; } || echo "$cd_lane_execstart" | grep -q "/bin/false"; }; then
cd_lane_ok=1
cd_lane_mode=failclosed
elif [ "$cd_lane_sentinel" = "present" ] && [ "$cd_lane_active" = "active" ] && [ "$cd_lane_capacity_ok" = "1" ] && [ "$cd_lane_labels_ok" = "1" ] && [ "$cd_lane_binary_elf" = "1" ]; then
cd_lane_ok=1
cd_lane_mode=controlled_open
fi
echo "CD_LANE_CONTROLLED mode=$cd_lane_mode load=$cd_lane_load unitfile=$cd_lane_unitfile active=$cd_lane_active mainpid=$cd_lane_mainpid sentinel=$cd_lane_sentinel capacity=$cd_lane_capacity_ok labels=$cd_lane_labels_ok binary_elf=$cd_lane_binary_elf process_count=$cd_lane_process_count ok=$cd_lane_ok"
cd_lane_drain_load=$(systemctl show awoooi-cd-lane-drain.service -p LoadState --value 2>/dev/null || true)
cd_lane_drain_unitfile=$(systemctl show awoooi-cd-lane-drain.service -p UnitFileState --value 2>/dev/null || true)
cd_lane_drain_active=$(systemctl show awoooi-cd-lane-drain.service -p ActiveState --value 2>/dev/null || true)
cd_lane_drain_mainpid=$(systemctl show awoooi-cd-lane-drain.service -p MainPID --value 2>/dev/null || true)
cd_lane_drain_cpu_accounting=$(systemctl show awoooi-cd-lane-drain.service -p CPUAccounting --value 2>/dev/null || true)
cd_lane_drain_cpu_quota=$(systemctl show awoooi-cd-lane-drain.service -p CPUQuotaPerSecUSec --value 2>/dev/null || true)
cd_lane_drain_memory_accounting=$(systemctl show awoooi-cd-lane-drain.service -p MemoryAccounting --value 2>/dev/null || true)
cd_lane_drain_memory_max=$(systemctl show awoooi-cd-lane-drain.service -p MemoryMax --value 2>/dev/null || true)
cd_lane_drain_tasks_accounting=$(systemctl show awoooi-cd-lane-drain.service -p TasksAccounting --value 2>/dev/null || true)
cd_lane_drain_tasks_max=$(systemctl show awoooi-cd-lane-drain.service -p TasksMax --value 2>/dev/null || true)
cd_lane_drain_limits_ok=0
if [ "$cd_lane_drain_cpu_accounting" = "yes" ] \
&& [ -n "$cd_lane_drain_cpu_quota" ] && [ "$cd_lane_drain_cpu_quota" != "infinity" ] \
&& [ "$cd_lane_drain_memory_accounting" = "yes" ] \
&& [ -n "$cd_lane_drain_memory_max" ] && [ "$cd_lane_drain_memory_max" != "infinity" ] \
&& [ "$cd_lane_drain_tasks_accounting" = "yes" ] \
&& [ -n "$cd_lane_drain_tasks_max" ] && [ "$cd_lane_drain_tasks_max" != "infinity" ]; then
cd_lane_drain_limits_ok=1
fi
cd_lane_drain_capacity_ok=0
cd_lane_drain_labels_ok=0
if grep -Eq "^[[:space:]]+capacity:[[:space:]]*1[[:space:]]*$" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null; then
cd_lane_drain_capacity_ok=1
fi
if grep -q "awoooi-ubuntu:docker://192.168.0.110:5000/awoooi/ci-runner:act-22.04" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null \
&& grep -q "awoooi-host:host" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null \
&& ! grep -Eq "^[[:space:]]+- \".*(ubuntu-latest|stockplatform|headless|playwright)" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null; then
cd_lane_drain_labels_ok=1
fi
cd_lane_drain_binary_kind=$(file -b /home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled 2>/dev/null || echo missing)
cd_lane_drain_binary_elf=0
echo "$cd_lane_drain_binary_kind" | grep -qi "ELF" && cd_lane_drain_binary_elf=1
cd_lane_drain_process_count=$(pgrep -f "^/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled" 2>/dev/null | wc -l | tr -d " ")
cd_lane_drain_ok=0
cd_lane_drain_mode=blocked
if [ "$cd_lane_drain_active" != "active" ] \
&& [ "$cd_lane_drain_binary_elf" = "0" ] \
&& [ "$cd_lane_drain_process_count" = "0" ] \
&& { [ "$cd_lane_drain_load" = "not-found" ] || { [ "$cd_lane_drain_load" = "masked" ] && [ "$cd_lane_drain_unitfile" = "masked" ]; }; }; then
cd_lane_drain_ok=1
cd_lane_drain_mode=failclosed
elif [ "$cd_lane_drain_active" = "active" ] \
&& [ "$cd_lane_drain_capacity_ok" = "1" ] \
&& [ "$cd_lane_drain_labels_ok" = "1" ] \
&& [ "$cd_lane_drain_binary_elf" = "1" ] \
&& [ "$cd_lane_drain_limits_ok" = "1" ]; then
cd_lane_drain_ok=1
cd_lane_drain_mode=controlled_open
fi
echo "CD_LANE_DRAIN_CONTROLLED mode=$cd_lane_drain_mode load=$cd_lane_drain_load unitfile=$cd_lane_drain_unitfile active=$cd_lane_drain_active mainpid=$cd_lane_drain_mainpid capacity=$cd_lane_drain_capacity_ok labels=$cd_lane_drain_labels_ok binary_elf=$cd_lane_drain_binary_elf limits=$cd_lane_drain_limits_ok process_count=$cd_lane_drain_process_count ok=$cd_lane_drain_ok"
cd_lane_root_restore_left=unknown
if sudo -n true >/dev/null 2>&1; then
cd_lane_root_restore_left=$(sudo -n find /root -maxdepth 1 -type d \( -name "awoooi-runner-restore-sources-disabled*" -o -name "awoooi-cd-lane-disabled*" -o -name "awoooi-cd-lane-drain-disabled*" \) -print 2>/dev/null | wc -l | tr -d " ")
cd_lane_root_restore_left=$(sudo -n find /root -maxdepth 1 -type d \( -name "awoooi-cd-lane-disabled-*" -o -name "awoooi-cd-lane-drain-disabled-*" \) -print 2>/dev/null | wc -l | tr -d " ")
fi
echo "CD_LANE_ROOT_RESTORE_SOURCES left=$cd_lane_root_restore_left"
sentinel_left=0
for s in /run/awoooi-runner-host-enabled /run/awoooi-start-controlled-cd-lane /run/awoooi-start-controlled-cd-lane-drain /run/awoooi-start-cd-lane-allowed /run/awoooi-cd-lane-drain-ok /run/awoooi-cd-lane-ok /run/awoooi-cd-lane-enabled /run/awoooi-cd-lane-controlled-open; do
[ -e "$s" ] && sentinel_left=$((sentinel_left + 1))
done
echo "RUNNER_SENTINELS_LEFT $sentinel_left"
active_job_containers=$(docker ps --format "{{.Names}}" 2>/dev/null | grep -Ec "^(GITEA-ACTIONS-|awoooi-cd-)" || true)
echo "ACTIVE_JOB_CONTAINERS $active_job_containers"
cd_lane_guard_ok=0
if [ "$enforcer_timer_active" = "active" ] \
&& [ "$enforcer_timer_enabled" = "enabled" ] \
&& [ "$enforcer_service_result" = "success" ] \
&& [ "$authority_timer_active" = "active" ] \
&& [ "$authority_timer_enabled" = "enabled" ] \
&& [ "$authority_service_result" = "success" ] \
&& [ "$cd_lane_process_count" = "0" ] \
&& [ "$cd_lane_root_restore_left" = "0" ] \
&& [ "$sentinel_left" = "0" ] \
&& [ "$active_job_containers" = "0" ]; then
if { [ "$cd_lane_ok" = "1" ] || [ "$cd_lane_drain_ok" = "1" ]; } && [ "$cd_lane_root_restore_left" = "0" ]; then
cd_lane_guard_ok=1
fi
echo "CD_LANE_GUARDRAILS_OK $cd_lane_guard_ok"
direct_runner_count=$(pgrep -f "^/home/wooo/act-runner/act_runner|^/home/wooo/act-runner-controlled/act_runner|^/home/wooo/awoooi-controlled-runner/awoooi_controlled_runner" 2>/dev/null | wc -l | tr -d " ")
echo "RUNNER_DIRECT_PROCESS_COUNT $direct_runner_count"
for p in /home/wooo/awoooi-cd-lane/awoooi_cd_lane /home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled /home/wooo/act-runner/act_runner /home/wooo/act-runner/act_runner.real-20260628-runner-pressure-guard /home/wooo/act-runner-controlled/act_runner /home/wooo/awoooi-controlled-runner/awoooi_controlled_runner; do
for p in /home/wooo/act-runner/act_runner /home/wooo/act-runner/act_runner.real-20260628-runner-pressure-guard /home/wooo/act-runner-controlled/act_runner /home/wooo/awoooi-controlled-runner/awoooi_controlled_runner; do
kind=$(file -b "$p" 2>/dev/null || echo missing)
echo "RUNNER_FAILCLOSED_BINARY $p kind=$kind"
echo "$kind" | grep -qi "ELF" && echo "RUNNER_FAILCLOSED_BINARY_ELF $p"
@@ -369,15 +423,12 @@ docker ps --format "DOCKER {{.Names}}\t{{.Status}}" | head -120
warn "runner watchdog state not confirmed"
fi
if awk '$1 == "RUNNER_FAILCLOSED_UNIT" && $NF != "ok=1" {bad=1} END {exit bad}' <<<"$out"; then
ok "110 runner/CD lane units are fail-closed"
ok "110 legacy direct/Gitea runner units are fail-closed"
else
fail "110 runner/CD lane units are not fail-closed"
fail "110 legacy direct/Gitea runner units are not fail-closed"
fi
grep -q "RUNNER_FAILCLOSED_ENFORCER timer_active=active timer_enabled=enabled service_result=success" <<<"$out" && ok "110 fail-closed enforcer timer active and successful" || fail "110 fail-closed enforcer timer not healthy"
grep -q "RUNNER_FAILCLOSED_AUTHORITY timer_active=active timer_enabled=enabled service_result=success" <<<"$out" && ok "110 fail-closed authority timer active and successful" || fail "110 fail-closed authority timer not healthy"
grep -q "CD_LANE_GUARDRAILS_OK 1" <<<"$out" && ok "110 cd-lane/drain lane are fail-closed with enforcer" || fail "110 cd-lane/drain lane fail-closed guardrails incomplete"
grep -q "CD_LANE_GUARDRAILS_OK 1" <<<"$out" && ok "110 controlled cd-lane is safe, drained, or fail-closed" || fail "110 controlled cd-lane is neither safe-open/drained nor fail-closed"
grep -q "RUNNER_DIRECT_PROCESS_COUNT 0" <<<"$out" && ok "110 legacy direct runner process count is zero" || fail "110 legacy direct runner process detected"
grep -q "ACTIVE_JOB_CONTAINERS 0" <<<"$out" && ok "110 Gitea/CD job container count is zero" || fail "110 Gitea/CD job container still active"
grep -q "RUNNER_FAILCLOSED_BINARY_ELF" <<<"$out" && fail "110 runner fail-closed binary path restored to ELF" || ok "110 runner binary paths are fail-closed stubs or missing"
grep -q "sentry-self-hosted-clickhouse-1.*Restarting" <<<"$out" && warn "Sentry ClickHouse restarting" || ok "Sentry ClickHouse not visibly restarting"
}

View File

@@ -306,82 +306,137 @@ check_runner_guardrails() {
local out bad
if ! out=$(ssh_cmd "wooo@192.168.0.110" '
bad=0
for u in awoooi-cd-lane.service awoooi-cd-lane-drain.service awoooi-direct-runner-open.service awoooi-direct-runner.service gitea-act-runner-host.service gitea-act-runner-awoooi-controlled.service gitea-awoooi-controlled-runner.service gitea-act-runner-awoooi-open.service; do
for u in awoooi-direct-runner-open.service awoooi-direct-runner.service gitea-act-runner-host.service gitea-act-runner-awoooi-controlled.service gitea-awoooi-controlled-runner.service gitea-act-runner-awoooi-open.service; do
load=$(systemctl show "$u" -p LoadState --value 2>/dev/null || true)
unitfile=$(systemctl show "$u" -p UnitFileState --value 2>/dev/null || true)
active=$(systemctl show "$u" -p ActiveState --value 2>/dev/null || true)
mainpid=$(systemctl show "$u" -p MainPID --value 2>/dev/null || true)
unit_ok=0
unit_stub=0
if [ "$load" = "masked" ] && [ "$unitfile" = "masked" ] && [ "$active" = "inactive" ] && [ "${mainpid:-0}" = "0" ]; then
unit_ok=1
elif [ "$active" = "inactive" ] && [ "${mainpid:-0}" = "0" ] \
&& systemctl cat "$u" 2>/dev/null | grep -q "ConditionPathExists=/run/awoooi-runner-migrated-or-hard-limited"; then
unit_stub=1
if [ "$load" = "masked" ] && [ "$unitfile" = "masked" ] && [ "$active" = "inactive" ]; then
unit_ok=1
fi
echo "RUNNER_FAILCLOSED_UNIT $u load=$load unitfile=$unitfile active=$active mainpid=$mainpid stub=$unit_stub ok=$unit_ok"
echo "RUNNER_FAILCLOSED_UNIT $u load=$load unitfile=$unitfile active=$active ok=$unit_ok"
[ "$unit_ok" = "1" ] || bad=1
done
enforcer_timer_active=$(systemctl is-active awoooi-runner-failclosed-enforcer.timer 2>/dev/null || true)
enforcer_timer_enabled=$(systemctl is-enabled awoooi-runner-failclosed-enforcer.timer 2>/dev/null || true)
enforcer_service_result=$(systemctl show awoooi-runner-failclosed-enforcer.service -p Result --value 2>/dev/null || true)
echo "RUNNER_FAILCLOSED_ENFORCER timer_active=$enforcer_timer_active timer_enabled=$enforcer_timer_enabled service_result=$enforcer_service_result"
[ "$enforcer_timer_active" = "active" ] && [ "$enforcer_timer_enabled" = "enabled" ] && [ "$enforcer_service_result" = "success" ] || bad=1
authority_timer_active=$(systemctl is-active awoooi-runner-failclosed-authority.timer 2>/dev/null || true)
authority_timer_enabled=$(systemctl is-enabled awoooi-runner-failclosed-authority.timer 2>/dev/null || true)
authority_service_result=$(systemctl show awoooi-runner-failclosed-authority.service -p Result --value 2>/dev/null || true)
echo "RUNNER_FAILCLOSED_AUTHORITY timer_active=$authority_timer_active timer_enabled=$authority_timer_enabled service_result=$authority_service_result"
[ "$authority_timer_active" = "active" ] && [ "$authority_timer_enabled" = "enabled" ] && [ "$authority_service_result" = "success" ] || bad=1
cd_lane_process_count=$(pgrep -f "^/home/wooo/awoooi-cd-lane/awoooi_cd_lane|^/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled" 2>/dev/null | wc -l | tr -d " ")
echo "CD_LANE_PROCESS_COUNT $cd_lane_process_count"
[ "$cd_lane_process_count" = "0" ] || bad=1
cd_lane_load=$(systemctl show awoooi-cd-lane.service -p LoadState --value 2>/dev/null || true)
cd_lane_unitfile=$(systemctl show awoooi-cd-lane.service -p UnitFileState --value 2>/dev/null || true)
cd_lane_active=$(systemctl show awoooi-cd-lane.service -p ActiveState --value 2>/dev/null || true)
cd_lane_execstart=$(systemctl show awoooi-cd-lane.service -p ExecStart --value 2>/dev/null || true)
cd_lane_sentinel=missing
[ -e /run/awoooi-cd-lane-enabled ] && cd_lane_sentinel=present
cd_lane_capacity_ok=0
cd_lane_labels_ok=0
if grep -Eq "^[[:space:]]+capacity:[[:space:]]*1[[:space:]]*$" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null; then
cd_lane_capacity_ok=1
fi
if grep -q "awoooi-ubuntu:docker://192.168.0.110:5000/awoooi/ci-runner:act-22.04" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null \
&& grep -q "awoooi-host:host" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null \
&& ! grep -Eq "^[[:space:]]+- \".*(ubuntu-latest|stockplatform|headless|playwright)" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null; then
cd_lane_labels_ok=1
fi
cd_lane_binary_kind=$(file -b /home/wooo/awoooi-cd-lane/awoooi_cd_lane 2>/dev/null || echo missing)
cd_lane_binary_elf=0
echo "$cd_lane_binary_kind" | grep -qi "ELF" && cd_lane_binary_elf=1
cd_lane_process_count=$(pgrep -f "^/home/wooo/awoooi-cd-lane/awoooi_cd_lane" 2>/dev/null | wc -l | tr -d " ")
cd_lane_ok=0
cd_lane_mode=blocked
if [ "$cd_lane_active" = "inactive" ] \
&& [ "$cd_lane_sentinel" = "missing" ] \
&& [ "$cd_lane_binary_elf" = "0" ] \
&& [ "$cd_lane_process_count" = "0" ] \
&& { { [ "$cd_lane_load" = "masked" ] && [ "$cd_lane_unitfile" = "masked" ]; } || echo "$cd_lane_execstart" | grep -q "/bin/false"; }; then
cd_lane_ok=1
cd_lane_mode=failclosed
elif [ "$cd_lane_sentinel" = "present" ] && [ "$cd_lane_active" = "active" ] && [ "$cd_lane_capacity_ok" = "1" ] && [ "$cd_lane_labels_ok" = "1" ] && [ "$cd_lane_binary_elf" = "1" ]; then
cd_lane_ok=1
cd_lane_mode=controlled_open
fi
echo "CD_LANE_CONTROLLED mode=$cd_lane_mode load=$cd_lane_load unitfile=$cd_lane_unitfile active=$cd_lane_active sentinel=$cd_lane_sentinel capacity=$cd_lane_capacity_ok labels=$cd_lane_labels_ok binary_elf=$cd_lane_binary_elf process_count=$cd_lane_process_count ok=$cd_lane_ok"
cd_lane_drain_load=$(systemctl show awoooi-cd-lane-drain.service -p LoadState --value 2>/dev/null || true)
cd_lane_drain_unitfile=$(systemctl show awoooi-cd-lane-drain.service -p UnitFileState --value 2>/dev/null || true)
cd_lane_drain_active=$(systemctl show awoooi-cd-lane-drain.service -p ActiveState --value 2>/dev/null || true)
cd_lane_drain_cpu_accounting=$(systemctl show awoooi-cd-lane-drain.service -p CPUAccounting --value 2>/dev/null || true)
cd_lane_drain_cpu_quota=$(systemctl show awoooi-cd-lane-drain.service -p CPUQuotaPerSecUSec --value 2>/dev/null || true)
cd_lane_drain_memory_accounting=$(systemctl show awoooi-cd-lane-drain.service -p MemoryAccounting --value 2>/dev/null || true)
cd_lane_drain_memory_max=$(systemctl show awoooi-cd-lane-drain.service -p MemoryMax --value 2>/dev/null || true)
cd_lane_drain_tasks_accounting=$(systemctl show awoooi-cd-lane-drain.service -p TasksAccounting --value 2>/dev/null || true)
cd_lane_drain_tasks_max=$(systemctl show awoooi-cd-lane-drain.service -p TasksMax --value 2>/dev/null || true)
cd_lane_drain_limits_ok=0
if [ "$cd_lane_drain_cpu_accounting" = "yes" ] \
&& [ -n "$cd_lane_drain_cpu_quota" ] && [ "$cd_lane_drain_cpu_quota" != "infinity" ] \
&& [ "$cd_lane_drain_memory_accounting" = "yes" ] \
&& [ -n "$cd_lane_drain_memory_max" ] && [ "$cd_lane_drain_memory_max" != "infinity" ] \
&& [ "$cd_lane_drain_tasks_accounting" = "yes" ] \
&& [ -n "$cd_lane_drain_tasks_max" ] && [ "$cd_lane_drain_tasks_max" != "infinity" ]; then
cd_lane_drain_limits_ok=1
fi
cd_lane_drain_capacity_ok=0
cd_lane_drain_labels_ok=0
if grep -Eq "^[[:space:]]+capacity:[[:space:]]*1[[:space:]]*$" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null; then
cd_lane_drain_capacity_ok=1
fi
if grep -q "awoooi-ubuntu:docker://192.168.0.110:5000/awoooi/ci-runner:act-22.04" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null \
&& grep -q "awoooi-host:host" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null \
&& ! grep -Eq "^[[:space:]]+- \".*(ubuntu-latest|stockplatform|headless|playwright)" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null; then
cd_lane_drain_labels_ok=1
fi
cd_lane_drain_binary_kind=$(file -b /home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled 2>/dev/null || echo missing)
cd_lane_drain_binary_elf=0
echo "$cd_lane_drain_binary_kind" | grep -qi "ELF" && cd_lane_drain_binary_elf=1
cd_lane_drain_process_count=$(pgrep -f "^/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled" 2>/dev/null | wc -l | tr -d " ")
cd_lane_drain_ok=0
cd_lane_drain_mode=blocked
if [ "$cd_lane_drain_active" != "active" ] \
&& [ "$cd_lane_drain_binary_elf" = "0" ] \
&& [ "$cd_lane_drain_process_count" = "0" ] \
&& { [ "$cd_lane_drain_load" = "not-found" ] || { [ "$cd_lane_drain_load" = "masked" ] && [ "$cd_lane_drain_unitfile" = "masked" ]; }; }; then
cd_lane_drain_ok=1
cd_lane_drain_mode=failclosed
elif [ "$cd_lane_drain_active" = "active" ] \
&& [ "$cd_lane_drain_capacity_ok" = "1" ] \
&& [ "$cd_lane_drain_labels_ok" = "1" ] \
&& [ "$cd_lane_drain_binary_elf" = "1" ] \
&& [ "$cd_lane_drain_limits_ok" = "1" ]; then
cd_lane_drain_ok=1
cd_lane_drain_mode=controlled_open
fi
echo "CD_LANE_DRAIN_CONTROLLED mode=$cd_lane_drain_mode load=$cd_lane_drain_load unitfile=$cd_lane_drain_unitfile active=$cd_lane_drain_active capacity=$cd_lane_drain_capacity_ok labels=$cd_lane_drain_labels_ok binary_elf=$cd_lane_drain_binary_elf limits=$cd_lane_drain_limits_ok process_count=$cd_lane_drain_process_count ok=$cd_lane_drain_ok"
cd_lane_root_restore_left=unknown
if sudo -n true >/dev/null 2>&1; then
cd_lane_root_restore_left=$(sudo -n find /root -maxdepth 1 -type d \( -name "awoooi-runner-restore-sources-disabled*" -o -name "awoooi-cd-lane-disabled*" -o -name "awoooi-cd-lane-drain-disabled*" \) -print 2>/dev/null | wc -l | tr -d " ")
cd_lane_root_restore_left=$(sudo -n find /root -maxdepth 1 -type d \( -name "awoooi-cd-lane-disabled-*" -o -name "awoooi-cd-lane-drain-disabled-*" \) -print 2>/dev/null | wc -l | tr -d " ")
fi
echo "CD_LANE_ROOT_RESTORE_SOURCES left=$cd_lane_root_restore_left"
if [ "$cd_lane_root_restore_left" = "0" ]; then
:
else
bad=1
cd_lane_guard_ok=0
if { [ "$cd_lane_ok" = "1" ] || [ "$cd_lane_drain_ok" = "1" ]; } && [ "$cd_lane_root_restore_left" = "0" ]; then
cd_lane_guard_ok=1
fi
sentinel_left=0
for s in /run/awoooi-runner-host-enabled /run/awoooi-start-controlled-cd-lane /run/awoooi-start-controlled-cd-lane-drain /run/awoooi-start-cd-lane-allowed /run/awoooi-cd-lane-drain-ok /run/awoooi-cd-lane-ok /run/awoooi-cd-lane-enabled /run/awoooi-cd-lane-controlled-open; do
[ -e "$s" ] && sentinel_left=$((sentinel_left + 1))
done
echo "RUNNER_SENTINELS_LEFT $sentinel_left"
[ "$sentinel_left" = "0" ] || bad=1
echo "CD_LANE_GUARDRAILS_OK $cd_lane_guard_ok"
[ "$cd_lane_guard_ok" = "1" ] || bad=1
direct_runner_count=$(pgrep -f "^/home/wooo/act-runner/act_runner|^/home/wooo/act-runner-controlled/act_runner|^/home/wooo/awoooi-controlled-runner/awoooi_controlled_runner" 2>/dev/null | wc -l | tr -d " ")
echo "RUNNER_DIRECT_PROCESS_COUNT $direct_runner_count"
[ "$direct_runner_count" = "0" ] || bad=1
job_count=$(docker ps --format "{{.Names}}" 2>/dev/null | grep -Ec "^(GITEA-ACTIONS-|awoooi-cd-)" || true)
echo "ACTIVE_JOB_CONTAINERS $job_count"
[ "$job_count" = "0" ] || bad=1
for p in /home/wooo/awoooi-cd-lane/awoooi_cd_lane /home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled /home/wooo/act-runner/act_runner /home/wooo/act-runner/act_runner.real-20260628-runner-pressure-guard /home/wooo/act-runner-controlled/act_runner /home/wooo/awoooi-controlled-runner/awoooi_controlled_runner; do
for p in /home/wooo/act-runner/act_runner /home/wooo/act-runner/act_runner.real-20260628-runner-pressure-guard /home/wooo/act-runner-controlled/act_runner /home/wooo/awoooi-controlled-runner/awoooi_controlled_runner; do
kind=$(file -b "$p" 2>/dev/null || echo missing)
echo "RUNNER_FAILCLOSED_BINARY $p kind=$kind"
echo "$kind" | grep -qi "ELF" && bad=1
done
cd_lane_guard_ok=0
[ "$bad" = "0" ] && cd_lane_guard_ok=1
echo "CD_LANE_GUARDRAILS_OK $cd_lane_guard_ok"
for u in $(systemctl list-units "actions.runner.*" --all --no-legend --plain 2>/dev/null | awk "{print \$1}"); do
load=$(systemctl show "$u" -p LoadState --value)
unitfile=$(systemctl show "$u" -p UnitFileState --value)
mainpid=$(systemctl show "$u" -p MainPID --value)
watchdog=$(systemctl show "$u" -p WatchdogUSec --value)
quota=$(systemctl show "$u" -p CPUQuotaPerSecUSec --value)
memory=$(systemctl show "$u" -p MemoryMax --value)
state=$(systemctl show "$u" -p ActiveState --value)
unitfile=$(systemctl show "$u" -p UnitFileState --value)
echo "$u watchdog=$watchdog quota=$quota memory=$memory state=$state unitfile=$unitfile"
if [ "$state" = "active" ] || [ "$state" = "activating" ]; then
[ "$watchdog" = "0" ] || bad=1
[ "$quota" != "infinity" ] && [ "$quota" != "0" ] || bad=1
[ "$memory" != "infinity" ] && [ "$memory" != "0" ] || bad=1
elif [ "$unitfile" = "masked" ] || [ "$state" = "inactive" ]; then
:
else
bad=1
action_ok=0
action_mode=blocked
if [ "$state" != "active" ] \
&& { [ "$load" = "masked" ] || [ "$load" = "not-found" ] || [ "$unitfile" = "masked" ] || [ "$unitfile" = "disabled" ]; } \
&& [ "${mainpid:-0}" = "0" ]; then
action_ok=1
action_mode=github_disabled
fi
echo "$u mode=$action_mode load=$load unitfile=$unitfile state=$state mainpid=$mainpid watchdog=$watchdog quota=$quota memory=$memory ok=$action_ok"
[ "$action_ok" = "1" ] || bad=1
done
echo "BAD_RUNNER_GUARDRAILS $bad"
' 2>&1); then
@@ -390,7 +445,7 @@ echo "BAD_RUNNER_GUARDRAILS $bad"
return
fi
echo "$out"
grep -q "BAD_RUNNER_GUARDRAILS 0" <<<"$out" && ok "110 runner/CD lane fail-closed enforcer and guardrails complete" || blocked "110 runner/CD lane fail-closed guardrails incomplete"
grep -q "BAD_RUNNER_GUARDRAILS 0" <<<"$out" && ok "legacy runner fail-closed and controlled cd-lane guardrails complete" || blocked "legacy runner / controlled cd-lane guardrails incomplete"
}
check_job_containers() {

View File

@@ -538,61 +538,112 @@ fi
section "110 runner fail-closed guard"
runner_tmp="$(mktemp -t post-start-runner.XXXXXX)"
if ssh_read "wooo@192.168.0.110" '
for u in awoooi-cd-lane.service awoooi-cd-lane-drain.service awoooi-direct-runner-open.service awoooi-direct-runner.service gitea-act-runner-host.service gitea-act-runner-awoooi-controlled.service gitea-awoooi-controlled-runner.service gitea-act-runner-awoooi-open.service; do
for u in awoooi-direct-runner-open.service awoooi-direct-runner.service gitea-act-runner-host.service gitea-act-runner-awoooi-controlled.service gitea-awoooi-controlled-runner.service gitea-act-runner-awoooi-open.service; do
load=$(systemctl show "$u" -p LoadState --value 2>/dev/null || true)
unitfile=$(systemctl show "$u" -p UnitFileState --value 2>/dev/null || true)
active=$(systemctl show "$u" -p ActiveState --value 2>/dev/null || true)
mainpid=$(systemctl show "$u" -p MainPID --value 2>/dev/null || true)
unit_ok=0
unit_stub=0
if [ "$load" = "masked" ] && [ "$unitfile" = "masked" ] && [ "$active" = "inactive" ] && [ "${mainpid:-0}" = "0" ]; then
unit_ok=1
elif [ "$active" = "inactive" ] && [ "${mainpid:-0}" = "0" ] \
&& systemctl cat "$u" 2>/dev/null | grep -q "ConditionPathExists=/run/awoooi-runner-migrated-or-hard-limited"; then
unit_stub=1
if [ "$load" = "masked" ] && [ "$unitfile" = "masked" ] && [ "$active" = "inactive" ]; then
unit_ok=1
fi
echo "RUNNER_FAILCLOSED_UNIT $u load=$load unitfile=$unitfile active=$active mainpid=$mainpid stub=$unit_stub ok=$unit_ok"
echo "RUNNER_FAILCLOSED_UNIT $u load=$load unitfile=$unitfile active=$active mainpid=$mainpid ok=$unit_ok"
done
enforcer_timer_active=$(systemctl is-active awoooi-runner-failclosed-enforcer.timer 2>/dev/null || true)
enforcer_timer_enabled=$(systemctl is-enabled awoooi-runner-failclosed-enforcer.timer 2>/dev/null || true)
enforcer_service_result=$(systemctl show awoooi-runner-failclosed-enforcer.service -p Result --value 2>/dev/null || true)
echo "RUNNER_FAILCLOSED_ENFORCER timer_active=$enforcer_timer_active timer_enabled=$enforcer_timer_enabled service_result=$enforcer_service_result"
authority_timer_active=$(systemctl is-active awoooi-runner-failclosed-authority.timer 2>/dev/null || true)
authority_timer_enabled=$(systemctl is-enabled awoooi-runner-failclosed-authority.timer 2>/dev/null || true)
authority_service_result=$(systemctl show awoooi-runner-failclosed-authority.service -p Result --value 2>/dev/null || true)
echo "RUNNER_FAILCLOSED_AUTHORITY timer_active=$authority_timer_active timer_enabled=$authority_timer_enabled service_result=$authority_service_result"
cd_lane_process_count=$(pgrep -f "^/home/wooo/awoooi-cd-lane/awoooi_cd_lane|^/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled" 2>/dev/null | wc -l | tr -d " ")
echo "CD_LANE_PROCESS_COUNT $cd_lane_process_count"
cd_lane_load=$(systemctl show awoooi-cd-lane.service -p LoadState --value 2>/dev/null || true)
cd_lane_unitfile=$(systemctl show awoooi-cd-lane.service -p UnitFileState --value 2>/dev/null || true)
cd_lane_active=$(systemctl show awoooi-cd-lane.service -p ActiveState --value 2>/dev/null || true)
cd_lane_mainpid=$(systemctl show awoooi-cd-lane.service -p MainPID --value 2>/dev/null || true)
cd_lane_execstart=$(systemctl show awoooi-cd-lane.service -p ExecStart --value 2>/dev/null || true)
cd_lane_sentinel=missing
[ -e /run/awoooi-cd-lane-enabled ] && cd_lane_sentinel=present
cd_lane_capacity_ok=0
cd_lane_labels_ok=0
if grep -Eq "^[[:space:]]+capacity:[[:space:]]*1[[:space:]]*$" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null; then
cd_lane_capacity_ok=1
fi
if grep -q "awoooi-ubuntu:docker://192.168.0.110:5000/awoooi/ci-runner:act-22.04" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null \
&& grep -q "awoooi-host:host" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null \
&& ! grep -Eq "^[[:space:]]+- \".*(ubuntu-latest|stockplatform|headless|playwright)" /home/wooo/awoooi-cd-lane/config.yaml 2>/dev/null; then
cd_lane_labels_ok=1
fi
cd_lane_binary_kind=$(file -b /home/wooo/awoooi-cd-lane/awoooi_cd_lane 2>/dev/null || echo missing)
cd_lane_binary_elf=0
echo "$cd_lane_binary_kind" | grep -qi "ELF" && cd_lane_binary_elf=1
cd_lane_process_count=$(pgrep -f "^/home/wooo/awoooi-cd-lane/awoooi_cd_lane" 2>/dev/null | wc -l | tr -d " ")
cd_lane_ok=0
cd_lane_mode=blocked
if [ "$cd_lane_active" = "inactive" ] \
&& [ "$cd_lane_sentinel" = "missing" ] \
&& [ "$cd_lane_binary_elf" = "0" ] \
&& [ "$cd_lane_process_count" = "0" ] \
&& { { [ "$cd_lane_load" = "masked" ] && [ "$cd_lane_unitfile" = "masked" ]; } || echo "$cd_lane_execstart" | grep -q "/bin/false"; }; then
cd_lane_ok=1
cd_lane_mode=failclosed
fi
echo "CD_LANE_CONTROLLED mode=$cd_lane_mode load=$cd_lane_load unitfile=$cd_lane_unitfile active=$cd_lane_active mainpid=$cd_lane_mainpid sentinel=$cd_lane_sentinel capacity=$cd_lane_capacity_ok labels=$cd_lane_labels_ok binary_elf=$cd_lane_binary_elf process_count=$cd_lane_process_count ok=$cd_lane_ok"
cd_lane_drain_load=$(systemctl show awoooi-cd-lane-drain.service -p LoadState --value 2>/dev/null || true)
cd_lane_drain_unitfile=$(systemctl show awoooi-cd-lane-drain.service -p UnitFileState --value 2>/dev/null || true)
cd_lane_drain_active=$(systemctl show awoooi-cd-lane-drain.service -p ActiveState --value 2>/dev/null || true)
cd_lane_drain_mainpid=$(systemctl show awoooi-cd-lane-drain.service -p MainPID --value 2>/dev/null || true)
cd_lane_drain_cpu_accounting=$(systemctl show awoooi-cd-lane-drain.service -p CPUAccounting --value 2>/dev/null || true)
cd_lane_drain_cpu_quota=$(systemctl show awoooi-cd-lane-drain.service -p CPUQuotaPerSecUSec --value 2>/dev/null || true)
cd_lane_drain_memory_accounting=$(systemctl show awoooi-cd-lane-drain.service -p MemoryAccounting --value 2>/dev/null || true)
cd_lane_drain_memory_max=$(systemctl show awoooi-cd-lane-drain.service -p MemoryMax --value 2>/dev/null || true)
cd_lane_drain_tasks_accounting=$(systemctl show awoooi-cd-lane-drain.service -p TasksAccounting --value 2>/dev/null || true)
cd_lane_drain_tasks_max=$(systemctl show awoooi-cd-lane-drain.service -p TasksMax --value 2>/dev/null || true)
cd_lane_drain_limits_ok=0
if [ "$cd_lane_drain_cpu_accounting" = "yes" ] \
&& [ -n "$cd_lane_drain_cpu_quota" ] && [ "$cd_lane_drain_cpu_quota" != "infinity" ] \
&& [ "$cd_lane_drain_memory_accounting" = "yes" ] \
&& [ -n "$cd_lane_drain_memory_max" ] && [ "$cd_lane_drain_memory_max" != "infinity" ] \
&& [ "$cd_lane_drain_tasks_accounting" = "yes" ] \
&& [ -n "$cd_lane_drain_tasks_max" ] && [ "$cd_lane_drain_tasks_max" != "infinity" ]; then
cd_lane_drain_limits_ok=1
fi
cd_lane_drain_capacity_ok=0
cd_lane_drain_labels_ok=0
if grep -Eq "^[[:space:]]+capacity:[[:space:]]*1[[:space:]]*$" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null; then
cd_lane_drain_capacity_ok=1
fi
if grep -q "awoooi-ubuntu:docker://192.168.0.110:5000/awoooi/ci-runner:act-22.04" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null \
&& grep -q "awoooi-host:host" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null \
&& ! grep -Eq "^[[:space:]]+- \".*(ubuntu-latest|stockplatform|headless|playwright)" /home/wooo/awoooi-cd-lane-drain/config.yaml 2>/dev/null; then
cd_lane_drain_labels_ok=1
fi
cd_lane_drain_binary_kind=$(file -b /home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled 2>/dev/null || echo missing)
cd_lane_drain_binary_elf=0
echo "$cd_lane_drain_binary_kind" | grep -qi "ELF" && cd_lane_drain_binary_elf=1
cd_lane_drain_process_count=$(pgrep -f "^/home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled" 2>/dev/null | wc -l | tr -d " ")
cd_lane_drain_ok=0
cd_lane_drain_mode=blocked
if [ "$cd_lane_drain_active" != "active" ] \
&& [ "$cd_lane_drain_binary_elf" = "0" ] \
&& [ "$cd_lane_drain_process_count" = "0" ] \
&& { [ "$cd_lane_drain_load" = "not-found" ] || { [ "$cd_lane_drain_load" = "masked" ] && [ "$cd_lane_drain_unitfile" = "masked" ]; }; }; then
cd_lane_drain_ok=1
cd_lane_drain_mode=failclosed
elif [ "$cd_lane_drain_active" = "active" ] \
&& [ "$cd_lane_drain_capacity_ok" = "1" ] \
&& [ "$cd_lane_drain_labels_ok" = "1" ] \
&& [ "$cd_lane_drain_binary_elf" = "1" ] \
&& [ "$cd_lane_drain_limits_ok" = "1" ]; then
cd_lane_drain_ok=1
cd_lane_drain_mode=controlled_open
fi
echo "CD_LANE_DRAIN_CONTROLLED mode=$cd_lane_drain_mode load=$cd_lane_drain_load unitfile=$cd_lane_drain_unitfile active=$cd_lane_drain_active mainpid=$cd_lane_drain_mainpid capacity=$cd_lane_drain_capacity_ok labels=$cd_lane_drain_labels_ok binary_elf=$cd_lane_drain_binary_elf limits=$cd_lane_drain_limits_ok process_count=$cd_lane_drain_process_count ok=$cd_lane_drain_ok"
cd_lane_root_restore_left=unknown
if sudo -n true >/dev/null 2>&1; then
cd_lane_root_restore_left=$(sudo -n find /root -maxdepth 1 -type d \( -name "awoooi-runner-restore-sources-disabled*" -o -name "awoooi-cd-lane-disabled*" -o -name "awoooi-cd-lane-drain-disabled*" \) -print 2>/dev/null | wc -l | tr -d " ")
cd_lane_root_restore_left=$(sudo -n find /root -maxdepth 1 -type d \( -name "awoooi-cd-lane-disabled-*" -o -name "awoooi-cd-lane-drain-disabled-*" \) -print 2>/dev/null | wc -l | tr -d " ")
fi
echo "CD_LANE_ROOT_RESTORE_SOURCES left=$cd_lane_root_restore_left"
sentinel_left=0
for s in /run/awoooi-runner-host-enabled /run/awoooi-start-controlled-cd-lane /run/awoooi-start-controlled-cd-lane-drain /run/awoooi-start-cd-lane-allowed /run/awoooi-cd-lane-drain-ok /run/awoooi-cd-lane-ok /run/awoooi-cd-lane-enabled /run/awoooi-cd-lane-controlled-open; do
[ -e "$s" ] && sentinel_left=$((sentinel_left + 1))
done
echo "RUNNER_SENTINELS_LEFT $sentinel_left"
active_job_containers=$(docker ps --format "{{.Names}}" 2>/dev/null | grep -Ec "^(GITEA-ACTIONS-|awoooi-cd-)" || true)
echo "ACTIVE_JOB_CONTAINERS $active_job_containers"
cd_lane_guard_ok=0
if [ "$enforcer_timer_active" = "active" ] \
&& [ "$enforcer_timer_enabled" = "enabled" ] \
&& [ "$enforcer_service_result" = "success" ] \
&& [ "$authority_timer_active" = "active" ] \
&& [ "$authority_timer_enabled" = "enabled" ] \
&& [ "$authority_service_result" = "success" ] \
&& [ "$cd_lane_process_count" = "0" ] \
&& [ "$cd_lane_root_restore_left" = "0" ] \
&& [ "$sentinel_left" = "0" ] \
&& [ "$active_job_containers" = "0" ]; then
if { [ "$cd_lane_ok" = "1" ] || [ "$cd_lane_drain_ok" = "1" ]; } && [ "$cd_lane_root_restore_left" = "0" ]; then
cd_lane_guard_ok=1
fi
echo "CD_LANE_GUARDRAILS_OK $cd_lane_guard_ok"
direct_runner_count=$(pgrep -f "^/home/wooo/act-runner/act_runner|^/home/wooo/act-runner-controlled/act_runner|^/home/wooo/awoooi-controlled-runner/awoooi_controlled_runner" 2>/dev/null | wc -l | tr -d " ")
echo "RUNNER_DIRECT_PROCESS_COUNT $direct_runner_count"
for p in /home/wooo/awoooi-cd-lane/awoooi_cd_lane /home/wooo/awoooi-cd-lane-drain/awoooi_cd_lane_controlled /home/wooo/act-runner/act_runner /home/wooo/act-runner/act_runner.real-20260628-runner-pressure-guard /home/wooo/act-runner-controlled/act_runner /home/wooo/awoooi-controlled-runner/awoooi_controlled_runner; do
for p in /home/wooo/act-runner/act_runner /home/wooo/act-runner/act_runner.real-20260628-runner-pressure-guard /home/wooo/act-runner-controlled/act_runner /home/wooo/awoooi-controlled-runner/awoooi_controlled_runner; do
kind=$(file -b "$p" 2>/dev/null || echo missing)
echo "RUNNER_FAILCLOSED_BINARY $p kind=$kind"
echo "$kind" | grep -qi "ELF" && echo "RUNNER_FAILCLOSED_BINARY_ELF $p"
@@ -606,15 +657,12 @@ else
fi
cat "$runner_tmp"
if awk '$1 == "RUNNER_FAILCLOSED_UNIT" && $NF != "ok=1" {bad=1} END {exit bad}' "$runner_tmp"; then
ok "110 runner/CD lane units are fail-closed"
ok "110 legacy direct/Gitea runner units are fail-closed"
else
blocked "110 runner/CD lane units are not fail-closed"
blocked "110 legacy direct/Gitea runner units are not fail-closed"
fi
grep -q "RUNNER_FAILCLOSED_ENFORCER timer_active=active timer_enabled=enabled service_result=success" "$runner_tmp" && ok "110 fail-closed enforcer timer active and successful" || blocked "110 fail-closed enforcer timer not healthy"
grep -q "RUNNER_FAILCLOSED_AUTHORITY timer_active=active timer_enabled=enabled service_result=success" "$runner_tmp" && ok "110 fail-closed authority timer active and successful" || blocked "110 fail-closed authority timer not healthy"
grep -q "CD_LANE_GUARDRAILS_OK 1" "$runner_tmp" && ok "110 cd-lane/drain lane are fail-closed with enforcer" || blocked "110 cd-lane/drain lane fail-closed guardrails incomplete"
grep -q "CD_LANE_GUARDRAILS_OK 1" "$runner_tmp" && ok "110 controlled cd-lane is safe-open/drained or fail-closed" || blocked "110 controlled cd-lane guardrails incomplete"
grep -q "RUNNER_DIRECT_PROCESS_COUNT 0" "$runner_tmp" && ok "110 legacy direct runner process count is zero" || blocked "110 legacy direct runner process detected"
grep -q "ACTIVE_JOB_CONTAINERS 0" "$runner_tmp" && ok "110 Gitea/CD job container count is zero" || blocked "110 Gitea/CD job container still active"
grep -q "RUNNER_FAILCLOSED_BINARY_ELF" "$runner_tmp" && blocked "110 runner fail-closed binary path restored to ELF" || ok "110 runner binary paths are fail-closed stubs or missing"
grep -q "RUNNER_PRESSURE_GATE_RC 0" "$runner_tmp" && ok "110 host pressure gate returned 0" || blocked "110 host pressure gate is blocking"
rm -f "$runner_tmp"