Track HiveGrid Cloud planning reference

This commit is contained in:
david committed 2026-06-26 20:48:01 +02:00
1 parent b3d5ea6764
commit e341c5938a
5784 files changed
+1104929 -3

No files matched your search

@@ -0,0 +1,80 @@
# BLOCKERS
> **READINESS_AUTHORITY: non-authoritative blocker register.** This file is a
> narrow list of unresolved external/operator risks and is not the deploy
> readiness source of truth. The canonical readiness source is
> `docs/improvement-plan/OPERATOR-CHECKLIST.md`; feature/runtime truth remains
> in the linked feature docs, manifests, and runbooks.
**Live external/operator blocker register.** Keep only unresolved risks here.
Resolved incident details are intentionally not kept in this live file; use git
history for forensic archaeology and the canonical feature/runbook docs for
current runtime truth.
## Active Blockers
### B-006 — External Security Review of WASM Sandbox
**Status:** waiting-for-operator-procurement
**Owner:** operator/external security reviewer
**Blocks:** production rollout of third-party plugin execution.
The plugin runtime foundation has internal tests and production fail-closed
gates, but production enablement remains blocked until an external security firm
reviews the WASM sandbox and capability gateway.
Review scope:
- Sandbox-escape resistance
- Capability-gateway tampering resistance
- Ed25519 signature trust model
- Crash-tracker bypass scenarios
- Audit-log tamper-evidence
Staging/foundation use is allowed behind `PLUGIN_RUNTIME_PRODUCTION_ENABLED=false`.
Production signoff must happen through
`docs/security/plugin-runtime-signoff.md`; current implementation truth lives in
`docs/features/plugins.md`, `docs/operations/plugin-lifecycle.md`, and
`docs/security/plugin-internal-findings.md`.
Next operator step: procure the review, file findings under
`docs/improvement-plan/06-plugin-system/security-findings/`, address all
Critical/High findings, then record final signoff in the security runbook.
## Unresolved Risks
### B-012 — Gitea-runner Docker `default-address-pool` recurrence risk
**Status:** mitigated-now-but-recurs-without-config-change
**Owner:** operator
**Blocks:** no active deploy; future push-to-main deploys can be blocked if
self-hosted runner Docker networks leak again.
The self-hosted Gitea Actions runner previously exhausted Docker's default
address pool after leaked per-job networks accumulated beside long-lived service
networks. A one-time prune restored deploy workflow execution, but recurrence
remains possible until the runner host is hardened.
Permanent operator actions:
1. Enlarge Docker `default-address-pools` on the runner host and restart Docker
during a maintenance window.
2. Add scheduled cleanup for orphaned Gitea Actions task networks.
3. Consider moving CI runners to dedicated infrastructure so CI network churn
cannot contend with long-lived service stacks.
Repo-side support already exists through deploy and Docker baseline checks:
`pnpm run verify:deploy-workflow` and `pnpm run verify:docker-baseline`.
## Resolved
Resolved entries were removed from this live register on 2026-05-28 to avoid
stale operational claims and sensitive incident residue. Canonical current docs:
- `docs/improvement-plan/OPERATOR-CHECKLIST.md`
- `docs/features/plugins.md`
- `docs/operations/plugin-lifecycle.md`
- `docs/ops/worker-queues.md`
- `docs/architecture/csrf.md`
- `docs/observability/sentry-selfhost.md`
- `docs/ops/stripe-meters-migration.md`