Skip to main content

FAQ

Common questions developers ask before integrating. For depth, follow the links to the relevant reference pages.

Plans & pricing​

Is there a free tier?​

Yes. The Free plan includes 50,000 jobs/month, 1,000 requests/hour, and 1 project. No credit card required. See pricing.

What happens if I reach my monthly job allowance?​

If opt-in overage is off (the default, and always the case on Free), POST /flare/v1/jobs returns 429 monthly_allowance_exceeded until the next billing period. If overage is on, jobs keep running and the extra volume is metered at your plan's per-10k rate — until your spend cap is hit, after which creates return 429 spend_cap_reached. In-flight jobs continue to process in all cases — only new enqueues are gated. Enabling overage or upgrading the plan unblocks immediately.

Do retries count against the monthly allowance?​

No. Only the first enqueue consumes a job from your allowance. Automatic retries (exponential backoff) and dead-letter movements do not increment the counter.

Can I pause my subscription?​

Stripe doesn't expose true subscription pause via the Customer Portal. To stop billing, cancel the subscription — your account reverts to the Free plan at the end of the current period. Re-subscribing later picks up where you left off; jobs and recurring schedules survive.

Do you offer non-profit / open-source discounts?​

Yes — contact sales. Send a one-paragraph project description and your usage estimate.

Data & privacy​

Where is my data stored?​

Customer data (jobs, payloads, audit logs) is stored in the Azure region your account was provisioned into. The default is the EU region; Enterprise customers can request a specific region (US, UK South, Australia East, etc.) at signup. See the Subprocessors page for the full vendor list and regions.

Is my data encrypted?​

Yes. Customer data is encrypted at rest (AES-256-GCM) and in transit (TLS 1.2+). API keys are stored as salted SHA-256 hashes; TOTP secrets and webhook signing secrets are encrypted at rest with per-environment keys.

Does Flare send my data to third parties?​

Only to the subprocessors named on the Subprocessors page and only for the stated purposes (payment processing, transactional email delivery, observability). No data is sold or used for advertising. Cookie usage is strictly-necessary only.

How do I get a GDPR data export?​

GET /platform/v1/account/export (JWT auth required) returns a streamed JSON file containing every job, audit-log row, and account record associated with your user. Rate limit: 1 export per 60-second window per user. See Account API § Export.

How do I delete my account?​

DELETE /platform/v1/account (JWT auth + password confirmation) hard-deletes the user, every project, every job, every audit-log row, every recurring schedule, and every alert setting. Your Stripe subscription is NOT cancelled by this endpoint — cancel it from the Stripe Customer Portal first. See Account API § Delete.

Do you have a DPA?​

Yes — contact sales to request a signed Data Processing Agreement. Standard Contractual Clauses (SCCs) for EU data transfers are available. See the DPA page for the customer-side form.

Reliability & retention​

What's the uptime SLA?​

99.9% monthly uptime for the API and job-processing pipeline on paid plans (Starter and above). The dashboard UI and marketing site are best-effort. See the SLA page for the credit table and measurement methodology, and the status page for live uptime.

How long are jobs retained?​

After a job reaches a terminal state (succeeded / failed / cancelled / dead_letter), it's retained for 30 days (Free / Starter), 90 days (Pro), or 180 days (Business) before being hard-deleted. Export jobs you need beyond that window via POST /flare/v1/jobs/export. See Limits & Quotas.

What happens if my webhook receiver is down?​

Flare retries up to 5 times with exponential backoff (1s → 2s → 4s → 8s → 16s + jitter), 10 seconds per attempt. After the final failed attempt, the delivery is marked failed. Retry decisions are based on HTTP status: 2xx = success, 4xx = permanent failure (no retry), 5xx / timeout / network error = retry. See Webhooks.

How do I make my jobs idempotent across retries?​

Send an Idempotency-Key header on the enqueue request. Same key + same body within 24 hours replays the original response. Same key + different body returns 409 idempotency_key_reuse. See the Idempotency guide.

Can I run Flare on-premises / self-hosted?​

Not today. Flare is a managed multi-tenant SaaS. If your compliance posture requires on-prem job processing, contact sales — we offer a single-tenant Enterprise deployment in a dedicated VPC.

Operations​

How do I monitor my workers and queue depth?​

Pull metrics from GET /flare/v1/metrics/summary, /throughput, /queues, and /job-types, or subscribe to job.dead_letter webhooks for incident alerts. See the Monitoring guide and Metrics API. For high-volume teams: contact sales for an OpenTelemetry export endpoint.

Can I run multiple workers behind a load balancer?​

Yes. Workers are stateless — each polls independently and the API claims jobs atomically. Just register every worker process (POST /flare/v1/workers/register) at startup and have it call /poll in a loop. See the Queues & concurrency guide.

How do I rotate my API key without dropping in-flight requests?​

POST /platform/v1/projects/{id}/keys/rotate issues a new key and keeps the old key valid for a 24-hour grace window. Update clients during the window, then revoke explicitly. See Authentication § Key rotation.

What happens during a Flare upgrade — do my jobs pause?​

No. Upgrades are zero-downtime: API hosts roll one at a time behind the load balancer; new requests route to live hosts; in-flight jobs continue on whatever worker leased them. Workers may briefly observe 503 service_unavailable from /poll during an upgrade window — retry with backoff.

Can I export the OpenAPI spec for use in Insomnia / Thunder Client?​

Yes — the current spec lives at https://docs.zeridion.com/api/openapi.json. Import it into any OpenAPI 3.x compatible tool.

See also​