Twilio status

How we check Twilio

We watch api.twilio.com from all five of our probe regions, every 2 minutes. A region that sees a failure does not make this page say "down" on its own: a second region has to agree first, the same confirmation rule every YoPingMe monitor uses.

We watch the API host every Twilio integration sends its requests to, and we pin the exact response an unauthenticated request gets rather than assuming any particular status code means health. The console and the marketing site are served separately and can both look fine while the API your application calls is failing.

What is Twilio?

Twilio turns telephony into an API call. Sending an SMS, placing a voice call, verifying a phone number, or running a two-factor flow becomes an HTTP request rather than a contract with a carrier, and that abstraction is why Twilio ended up underneath a large share of the messages and verification codes people receive without ever knowing whose infrastructure carried them.

The most important thing to understand about a Twilio problem is that "the API is up" and "my messages are arriving" are two different statements, and only the first one is something any status page can answer. When your application posts a message, Twilio accepts it and queues it. What happens after that involves mobile carriers, country routing rules, sender identity registration, and spam filtering, all outside Twilio's control and all capable of silently dropping traffic while every Twilio system reports itself healthy. A significant share of "Twilio is broken" incidents are actually a carrier filtering messages, a sender ID that was not registered for a particular country, or a regulatory change that took effect that week. The message logs in the Twilio console, with their per-message status and error codes, are the only place that distinguishes these, and they are far more informative than any green light.

The second thing worth knowing is how little Twilio's products share. Messaging, voice, email through SendGrid, video, and the verification products run on separate infrastructure with separate incident histories. An outage in one is not evidence about the others, which is why we watch the core API host rather than trying to summarise the whole company in a single dot. A page that averaged every Twilio product together would be reassuring at exactly the wrong moments.

Authentication errors deserve their own note, because they generate a disproportionate share of false alarms. A 401 from the API is not a failure of the platform, it is the platform working correctly and telling you the credential is wrong. Rotated auth tokens, deleted API keys, and requests aimed at the wrong subaccount all produce it, usually shortly after someone did some tidying up in the console. If an integration started failing at a time that correlates with a person being awake rather than with a traffic pattern, this is the first thing to check.

For applications where a message failing to arrive matters, the useful posture is to instrument delivery rather than availability. Twilio's status webhooks report the fate of each individual message, and a dashboard built on those tells you about a carrier problem hours before it would ever reach a public status page, if it reaches one at all.

Frequently asked questions

Is Twilio down right now?

Check the board above; it reflects our own probes against Twilio's API from all five of our regions, refreshed every couple of minutes. This measures whether the API responds, which is a narrower question than whether every message is being delivered.

My SMS messages are not arriving but Twilio looks up. Why?

Because delivery and API availability are different things, and this is the single most common Twilio confusion. Twilio accepting your request means the message entered the queue; whether it reaches a handset depends on downstream carriers, country-specific routing rules, sender registration, and filtering that Twilio does not control. Check the message status in your Twilio logs, since a message sitting at queued or undelivered with an error code tells you far more than any status page can.

Does a Twilio outage affect voice calls and SMS at the same time?

Not necessarily. Twilio runs voice, messaging, email, and its other products on infrastructure that fails independently, and its own incident history shows plenty of days where one product degraded while the rest were untouched. Check which components are affected rather than assuming an outage in one means an outage in all.

Why do I get a 401 from the Twilio API when nothing changed?

A 401 means Twilio answered and rejected the credential, so the service is working and the credential is not. The usual causes are an auth token that was rotated, an API key that was deleted, or a request that is authenticating against the wrong account SID after a subaccount change. None of these are outages, and none will appear on any status page.

What is the difference between this page and Twilio's own status page?

Twilio's page reports what their engineers have confirmed and published, broken down by product and region. This page reports what our probes measured from outside their network, on a fixed schedule, whether or not an update has been posted. Ours is the earlier and coarser signal; theirs is the authoritative and more detailed one.