coulomb/the-custodian
4
0
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity Actions

feat(workplan): CUST-WP-0067-T07 files repointed on both machines
All checks were successful
CI Smoke / host-smoke (push) Successful in 0s
Details
CI Smoke / container-smoke (push) Successful in 1s
Details

Browse source 
178 AGENTS.md files moved off the 127.0.0.1:18000 reverse tunnel to the
in-cluster address; both machines report zero stale rows. 100 pushed from the
workstation, 74 picked up by pull on the node, 4 handled individually.

binky-control committed but not pushed — it carries 6 pre-existing unpushed
commits that are not this change's to publish. activity-core was already 39
behind with an unrelated dirty file; only AGENTS.md was taken from origin.

state-hub-railiance01 is now unblocked for removal.

Still open: the block has no single source, so this recurs on the next
topology change; and 68 non-AGENTS.md references need individual judgement
rather than substitution.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
This commit is contained in:
codex 2026-08-25 00:25:07 +02:00
parent 8c76b65a66
commit 005d198bdd
2 changed files with 43 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

8
docs/recovery/tunnels-yaml-proposed-changes-CUST-WP-0067.md
View file

@ -53,10 +53,10 @@ Status 2026-08-24:
- `state-hub-mcp-railiance01` — **safe to remove now.** Central MCP is serving - `state-hub-mcp-railiance01` — **safe to remove now.** Central MCP is serving
and the global port map points at it. and the global port map points at it.
- `state-hub-railiance01` — **not yet.** The global instructions are repointed, - `state-hub-railiance01` — **safe to remove as of 2026-08-25.** All 178
but roughly 120 `AGENTS.md` files across both machines still tell agents to `AGENTS.md` files on both machines are repointed to the in-cluster address and
use `127.0.0.1:18000`. Removing it before `CUST-WP-0067-T07` repoints those both machines report zero stale rows. Nothing documented still points at this
breaks any session that follows its own repo's instructions. tunnel.
On railiance01 both services are reachable in-cluster with no tunnel at all — On railiance01 both services are reachable in-cluster with no tunnel at all —
this is where "abandon tunneling" genuinely applies. this is where "abandon tunneling" genuinely applies.

40
workplans/CUST-WP-0067-hub-authority-target-resolution.md
View file

@ -314,7 +314,7 @@ so authority depended on where a process happened to run.
```task ```task
id: CUST-WP-0067-T07 id: CUST-WP-0067-T07
status: todo status: progress
priority: high priority: high
state_hub_task_id: "dbc672d5-805a-567a-b548-4e8870e4fb2f" state_hub_task_id: "dbc672d5-805a-567a-b548-4e8870e4fb2f"
``` ```
@ -346,6 +346,44 @@ Acceptance: no `AGENTS.md` instructs a remote agent to use `127.0.0.1:18000`;
the block has one authoritative source; a spot check on both machines confirms the block has one authoritative source; a spot check on both machines confirms
the rendered guidance matches the live topology. the rendered guidance matches the live topology.
**Files repointed (2026-08-25).** Both machines now report zero stale rows.
178 files, one row each:
- **Workstation, 100 files** — all committed and pushed, `AGENTS.md` staged
alone so unrelated working-tree state was never swept into a commit.
- **Node, 78 files** — 74 picked the change up by `git pull` since they are
clones of the same origins. Four needed individual handling.
The block was near-uniform: 99 exact matches and one variant in
`whitehat-security` ("Remote tunnel", different table rule), which the pattern
covered. No file matched `18000` without being rewritten, so there was no
silent-skip tail. The `Local workstation` row was left alone — since `T02` it
is correct, because `127.0.0.1:8000` now *is* central.
Individual cases on the node:
| Repository | Condition | Action |
|---|---|---|
| `the-custodian` | untracked files only | pulled |
| `inter-hub` | origin never had the fix | edited, committed, pushed |
| `activity-core` | 39 behind, unrelated tracked file dirty | took `AGENTS.md` from origin only; pre-existing condition left for its owner |
| `binky-control` | 6 pre-existing unpushed commits | committed, **not pushed** — publishing someone else's work is not a side effect of a docs repoint |
**Remaining — the half that stops this recurring.** The block still has no single
source; it was copy-propagated by past sessions, and a grep for it finds only
the files themselves and old session logs. Until Repo Manager renders it (it
already owns `scaffold`), the next topology change is another 178-file pass.
**Also remaining, deliberately separate:** 68 non-`AGENTS.md` files still
reference `127.0.0.1:18000`, and they must not be blanket-rewritten —
21 scripts and code paths that actually configure things (`register-mcp.sh`,
`dev_hub_up.sh`), 21 docs and runbooks, and 26 workplans, reports and snapshots
that are historical record. Rewriting a finished workplan or a dated report
would falsify what was true at the time.
`state-hub-railiance01` is now unblocked for removal: nothing documented still
points at it.
## Deploy an MCP server on central ## Deploy an MCP server on central
```task ```task