From 02cfb27937a97fa4ed40bdb935481a7c52c1d8c9 Mon Sep 17 00:00:00 2001 From: Kyle Isom Date: Fri, 3 Apr 2026 09:33:41 -0700 Subject: [PATCH] Update incident action items: 4/7 complete Done: disaster recovery runbook, fallback DNS resolvers, sqlite chmod fix (mcdsl v1.8.0), UID stabilization. Remaining: agent recover command, boot sequencing, multi-address nodes. Co-Authored-By: Claude Opus 4.6 (1M context) --- log/2026-04-03-uid-incident.md | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-) diff --git a/log/2026-04-03-uid-incident.md b/log/2026-04-03-uid-incident.md index 5b558ea..930d99d 100644 --- a/log/2026-04-03-uid-incident.md +++ b/log/2026-04-03-uid-incident.md @@ -217,13 +217,14 @@ ownership changes. ## Action Items -- [ ] Write disaster recovery runbook (separate document) -- [ ] Add fallback DNS resolvers to NixOS config (1.1.1.1, 8.8.8.8) +- [x] Write disaster recovery runbook → `docs/disaster-recovery.md` +- [x] Add fallback DNS resolvers to NixOS config → all nodes now have + 1.1.1.1 and 8.8.8.8 as fallbacks after MCNS - [ ] Implement `mcp agent recover` command (recreate containers from registry) -- [ ] Implement boot sequencing in the agent (Phase 3+ architecture) -- [ ] Investigate modernc.org/sqlite `fchmod` workaround for rootless - podman +- [ ] Implement boot sequencing in the agent (Phase 4+ architecture) +- [x] Fix modernc.org/sqlite `fchmod` → was our own `os.Chmod` in + `mcdsl/db/db.go`, not sqlite. Made best-effort in mcdsl v1.8.0. - [ ] Add multi-address support to node config (Tailnet IP + LAN IP + hostname) -- [ ] Revert mcp UID to a stable value and never change it again +- [x] Stabilize mcp UID → pinned at 850 with NEVER CHANGE comment