FireFlo SMS/SMPP Gateway

Queues many messages in one call. The response confirms acceptance only — per-message results arrive on your errback URL.

{
  "messages": [
    { "to": "61491570006", "content": "Your OTP is 4417." },
    { "to": "61491570007", "content": "Your OTP is 9921." }
  ]
}

{
  "data": {
    "batchId": "c0ffee12-9f3a-4a1e-8d55-1b2c3d4e5f60",
    "messageCount": 3,
    "scheduled": "3600s"
  }
}

messageCount is how many the batch expanded to, not how many succeeded. There are no per-message ids in this response and no per-message results — see Batch callbacks.

Fields

Field	Type	Required	Notes
`messages`	array	yes	One object per message, each taking the fields of a single send. Its `to` may also be an array.
`globals`	object	no	Defaults merged under every message, so a field set on a message wins.
`batch_config`	object	no	`schedule_at`, `callback_url`, `errback_url`.

Other keys in batch_config are ignored, not refused. A misspelled errback_url costs you every per-message result while the batch still returns 200.This is the one place on the API where a typo does not produce a 400. Verify it by sending a batch containing one message you know will be refused.

`schedule_at`

Exactly two forms:

Form	Meaning
`3600s`	Relative — in one hour
`2026-08-20 09:30:00`	Absolute, `YYYY-MM-DD HH:MM:SS`

Anything else is 412 Got unknown format. A date in the past is refused, not sent immediately.

Errors particular to this endpoint

Status	`message`	When
400	`Mandatory argument messages is not found.`	Absent, not an array, or empty.
400	`Argument messages has an invalid value: a batch may expand to at most 10000 messages.`	Array-valued destinations multiplied out past the cap.
412	`Got unknown format: <value>, correct formats are …`	`schedule_at` in neither accepted form, or in the past.

The cap is 10 000 after expansion. 100 messages each with 200 recipients is 20 000, and refused.

Pacing and durability

Batches pace themselves against the router queue’s depth, pausing above a high-water mark — you do not need to throttle your own submission of a batch.

Scheduled batches survive a restart only with a RabbitMQ broker configured, and not even then while the broker connection is degraded. If durable scheduling matters, confirm with your provider which queue provider they run.

There is no batch status endpoint

Not in FireFlo and not in Jasmin. The batchId correlates the callbacks you receive; there is nothing to poll.

Batches

Worked examples and the pacing story.

Batch callbacks

Where per-message results actually arrive.

REST API

Configuration keys

File formats

SMPP settings

CLI

Management API

Data model

POST /secure/sendbatch

Fields

`schedule_at`

Errors particular to this endpoint

Pacing and durability

There is no batch status endpoint

Batches

Batch callbacks

​Fields

​schedule_at

​Errors particular to this endpoint

​Pacing and durability

​There is no batch status endpoint

​Related

Batches

Batch callbacks

Fields

`schedule_at`

Errors particular to this endpoint

Pacing and durability

There is no batch status endpoint

Related