docs: clarify current unifi protect chime state
Some checks failed
secret-guardrails / artifact-secret-scan (push) Has been cancelled
secret-guardrails / gitleaks (push) Has been cancelled

This commit is contained in:
Fizzlepoof
2026-05-26 20:10:45 +00:00
parent a282fe43bd
commit 77da65e64a
16 changed files with 193 additions and 43 deletions

View File

@@ -2,6 +2,10 @@
Purpose: turn the remaining SSID cleanup into an exact low-risk keep/retire proposal before any live Wi-Fi changes.
Status note:
- this remains useful as a keep/retire proposal, but the Protect chime portion must now be read together with `unifi-protect-doorbell-chime-current-state-2026-05-26.md`
- current known-good chime behavior is on Management/default `UniFi Wireless`, not on the Camera/Security lane
## Basis
Artifact-based planning only. No live SSID edits were made in this pass.
@@ -87,8 +91,11 @@ Long-term, the network should converge on:
- doorbell and Protect-chime estate justify a distinct security SSID
- keeping security separate from generic IoT matches the redesign intent and future camera growth plan
- Preconditions to call it final:
- doorbell/chimes remain healthy
- doorbell/chimes remain healthy after explicit Protect validation on that lane
- no evidence that merging them into general IoT would be safer or simpler
- Current caution:
- treat the Security SSID as future-state design intent, not proof that the Wi-Fi chimes are safe there today
- the current known-good live path for the chimes is still Management/default `UniFi Wireless`
### 5) Guest SSID
- Proposed role: `create or enable separately when ready`
@@ -116,6 +123,7 @@ This is the safest order based on current evidence:
- do not rename and repurpose a live SSID in one step if disabling/creating separately would be clearer
- do not infer that Google/cast is solved globally from one successful pilot reassociation
- do not collapse Security into generic IoT unless there is a deliberate design change
- do not assume that because `UNEF's Playhouse` exists, the Protect chimes are already proven safe on it
## Practical one-page operator version
If you want the shortest operator call: