Key Takeaways
- api how to incident response for outage is a runbook: radio vs control plane vs operator vs your worker — in that order.
- Most “API outages” are Doze, empty airtime, or unpaired phones. Check last-seen before paging the vendor.
- Pause bulk. Pin OTP to a spare. Idempotency-Key so retries do not double-send codes.
- We do not refund operator minutes burned during the incident.
- Priced by devices and SMS send volume. You use your own phone and operator SMS credit. A spare still meters as a device.
- App-side IR lives in the Android how-to. This spoke is the API worker.
Name the plane that failed
Search intent for api how to incident response for outage is what the API worker does when OTP stops. Service pricing is based on device count and total SMS sent through the gateway. Handset IR: Android incident response. Do not mix those runbooks.
If last-seen is six hours old, you do not have an API incident. You have a phone in a drawer.
NIST incident handling is a useful floor for roles and comms (SP 800-61). This is not a penetration guide.
Detect: last-seen, queue, DLR
Watch pairing last-seen, queue depth, and webhook message.failed — webhooks setup. GET /api/v1/messages/{id} for a single job. Docs: docs.sms-gateway.app.
| Plane | Symptom | First move |
|---|---|---|
| Android radio | Last-seen stale; queue grows | Power, Doze, pairing, SIM, spare |
| Your worker | Timeouts, duplicate OTPs | Idempotency-Key; stop the retry loop |
| Control plane | HTTP 5xx / auth across all devices | Status + docs; do not rewrite the client blindly |
| Operator | One SIM fails; others deliver | Balance, AUP, destination canary |
Contain: spare radio, pause bulk
Stop campaigns. Pin OTP with deviceIds to a charged spare. Failover troubleshooting. Device setup. Starter is 2 devices.
Do not retry-storm OTP
Ambiguous timeouts plus no Idempotency-Key duplicate codes. Keys stay in env — API keys in env. Rotate after a leaked key incident. OTP verification.
What you tell customers
Honest: codes delayed; do not promise a cloud SMSC SLA. Priced by devices and SMS send volume. You use your own phone and operator SMS credit. Airtime burned in retries is the operator’s bill.
IR checklist
- Plane named before the war room.
- Bulk paused; OTP pinned to spare.
- Retry loop stopped.
- Staff canary after recover.
- Postmortem: radio vs worker vs carrier.
Next steps
Drill: unplug primary, confirm spare OTP, restore. Pricing. App: downloads.
Related product pages
Jump to the live product docs for this topic—not another long-form article.
- SMS API documentationLive endpoint reference
- device and SMS volume pricingPlans and allowances
- Android SMS gateway product guideDefinition, product, and how to buy
- download the Android gateway appGet the APK
