UPES-ECS — Bare-Minimum Operational Checklist¶
The least you need to be genuinely operational — i.e. a student can dial 111, an ERT officer answers on a real device, the call is recorded, and an unanswered call becomes a reviewable incident. Two columns: Minimum (must have to go live) and Recommended (do this for a real, resilient pilot).
Rule of thumb: if any Minimum box is unchecked, you are not operational.
The 6 pillars¶
graph LR
S[1. Server<br/>Asterisk/FreePBX] --- N[2. Network<br/>Wi-Fi + LAN]
N --- P[3. Phones<br/>Android answer points]
P --- A[4. Accounts<br/>SAP IDs + positions]
A --- PE[5. People<br/>trained ERT on shift]
PE --- C[6. Config + Tests<br/>111 flow proven]
1. Server (the PBX)¶
| # | Item | Minimum | Recommended |
|---|---|---|---|
| 1.1 | Machine | 1× mini-PC / server (or the van PBX) | Dedicated, in a locked room/rack |
| 1.2 | OS | Ubuntu Server LTS / Debian stable | Same + auto-security-updates |
| 1.3 | PBX software | Asterisk 18+ / FreePBX (free) | FreePBX GUI for admin |
| 1.4 | Static IP | Mandatory — assigned + recorded | + local DNS pbx.upes.lan |
| 1.5 | Storage | Enough for recordings/logs (tens of GB) | Separate disk/partition for recordings |
| 1.6 | Power | UPS on the PBX (emergency system!) | UPS on PBX + switch + router + key APs |
⚑ Power is not optional for a disaster system. Minimum = UPS on the server so a brief outage doesn't kill 111. The van solves this with autonomous power.
2. Network¶
| # | Item | Minimum | Recommended |
|---|---|---|---|
| 2.1 | Wi-Fi AP | 1× AP covering the pilot area | APs covering all target zones |
| 2.2 | Switch + router | 1× each, LAN reachable to PBX | PoE switch + voice VLAN |
| 2.3 | Client isolation | OFF (or SIP/RTP allowed to PBX) | Voice-enabled SSID/VLAN |
| 2.4 | Firewall | SIP/RTP allowed LAN-only; no public exposure | + guest Wi-Fi blocked from PBX |
| 2.5 | Quality | Two-way audio works in pilot area | < 150 ms latency, < 1% loss |
3. Phones (answer points + callers)¶
| # | Item | Minimum | Recommended |
|---|---|---|---|
| 3.1 | ERT answer points | 3× dedicated Androids (ERT Lead 4101 + 2 operators 4110/4111) | 4–5 (add reserve + Control) |
| 3.2 | Answer-point care | On charger, battery-optimization OFF, screen-lock off | Wall-mounted, labelled per position |
| 3.3 | Responder devices | — (Medical/Security can start as dispatch-by-mobile) | Medical 4200 + Security 4300 Androids |
| 3.4 | Caller devices | Pilot users install Linphone on their own phones | 10–25 pilot users onboarded |
| 3.5 | App | Linphone (Android) | + MicroSIP for any Windows desk |
4. Accounts & identity¶
| # | Item | Minimum | Recommended |
|---|---|---|---|
| 4.1 | Responder positions | 4101, 4110, 4111 provisioned + registered |
+ 4112, 4113, 4120, 4200, 4300 |
| 4.2 | Pilot user accounts | ~10 SAP-ID accounts (pilot-users.csv) | 10–25 |
| 4.3 | Secrets | Unique, ≥12-char random per account | + delivered once, securely; stored in secrets store |
| 4.4 | Roles/contexts | ERT positions in ctx_ert/ctx_ert_lead; users in ctx_student/ctx_staff |
+ ctx_responder for Medical/Security |
| 4.5 | Anonymous SIP | Disabled; guest Wi-Fi blocked | + fail2ban |
5. People & process¶
| # | Item | Minimum | Recommended |
|---|---|---|---|
| 5.1 | ERT on shift | 2 operators available + 1 Lead reachable | 3+ operators, Lead + reserves |
| 5.2 | Shift roster | Who holds each position, per shift (SOP 30) | + handover log signed |
| 5.3 | IT admin | 1 who can run the PBX | 1 + a trained backup (bus factor) |
| 5.4 | Training | ERT know the answer script + dispatch (SOP 02) | Full Training Plan + a drill |
| 5.5 | Desk references | ERT SOP + Quick-Cards printed | + numbering + drill SOP |
6. Config, prompts & the go-live test¶
| # | Item | Minimum | Recommended |
|---|---|---|---|
| 6.1 | Emergency dialplan | 111 → queue → coach-in-parallel + background alert → voicemail live (config) | + paging/conference/dispatch |
| 6.2 | Recording | 111 recorded whole-call; files land + link to incident | + retention cron (SOP 13) |
| 6.3 | Incident logging | Missed 111 → Missed Emergency Incident | + full incident schema (SOP 12) |
| 6.4 | Prompts | emergency-preanswer, emergency-voicemail-prompt, drill-prompt recorded |
all prompts (SOP 28) |
| 6.5 | Health check | upes-ecs-healthcheck.sh runs, reports READY |
+ local dashboard |
| 6.6 | Backup | One tested config backup taken | daily + pre-change + tested restore (SOP 11) |
| 6.7 | Drill line | 199 works, no real dispatch | + drill schedule |
Go-live gate — these MUST pass (SOP 17 / 32)¶
- [ ] A registered softphone dials 111 → an ERT Android rings → answered.
- [ ] The call produces a recording linked to an incident ID.
- [ ] Unanswered 111 → coach-in-parallel + background alert → voicemail → Missed Emergency Incident (critical, pending).
- [ ] 199 drill runs with no real dispatch.
- [ ] Student-to-student call works and is not recorded.
- [ ] Health check reports READY (≥2 ERT positions available).
- [ ] A config backup exists and restore was tested once.
A failed 111 / 199 test or a recording failure is a do-not-go-live condition.
Absolute-minimum shopping list (if you have nothing but the network)¶
1 × mini-PC/server (Ubuntu) + 1 × UPS
3 × dedicated Android phones + 3 × chargers/wall mounts (ERT answer points)
(use existing router/switch/AP)
Free software: Asterisk/FreePBX, Linphone, this repo's config/scripts
People: 2 trained ERT operators + 1 Lead + 1 IT admin (per shift)
That, plus the config in ../config/ and the runbook, is enough to answer real emergencies on 111. Everything else is resilience and reach.
Full itemized kit (campus + van, minimum + recommended): 05-Bill-of-Materials.md.