RPF-WP-0015-T02: document driver reachability and restart gaps

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

Assistant: claude-code
Assistant-Model: opus
Assistant-Process: 680842@bnt-lap001
Assistant-Session: 54eb9e8e-7d6b-4f0c-b7a6-a6db62075883
This commit is contained in:
codex 2026-10-14 12:54:03 +02:00
parent 6152aa9e86
commit 8cfbf6410d
2 changed files with 9 additions and 1 deletions

View file

@ -111,6 +111,14 @@ driver as `AUDIT_T02_SOURCE`/`AUDIT_T02_TENANT`; the confirm string is
If cleanup is missed, audit-core's `expires_at` enforcement makes the identity
inert at window end; the attended cleanup is still required to remove it.
The driver runs operator-side. audit-core's NetworkPolicies admit only named
namespaces, so reach the pod with `kubectl -n audit-core port-forward
deploy/audit-core <local-port>:8080` and set `AUDIT_T02_BASE_URL` to that local
URL; record the base URL and the driver's source/tenant (which must equal the
receipt's `synthetic_sender` block) in the window notes. Projection and cleanup
each restart audit-core once (Recreate, no surge), so production senders see two
short retryable gaps per window; announce them with the window.
If `project` fails after the registry write may already have reached OpenBao (for
example a transport error on the write itself), treat the identity as possibly
present: run `cleanup --abort-reason "<reason>"` before leaving the window.