@kapa.ai Is there a way to set an alert for pipelines stalling or taking longer than usual to run?

Yes. There are a couple of options in Airbyte to alert you when syncs stall or behave abnormally: 1. Use built‑in notifications on sync completion/failure Airbyte can send notifications per workspace on several events, including: - Failed syncs - Successful syncs - Warning – Repeated Failures - Sync Disabled – Repeated Failures - Various connection updates and upgrade‑related events Notifications can be delivered via email (Cloud only) or webhook (Cloud and Self‑Managed).[[Manage notifications](https://docs.airbyte.com/platform/cloud/managing-airbyte-cloud/manage-airbyte-cloud-notifications)] You can wire the webhook to Slack or another system to get alerted when syncs fail or when Airbyte is about to/has disabled a sync because of repeated failures (which often indicates something is “stuck”).[[Webhook payloads](https://docs.airbyte.com/platform/cloud/managing-airbyte-cloud/manage-airbyte-cloud-notifications#set-up-webhook-notifications)] 2. Webhook payloads for long‑running/hanging jobs For each sync attempt, the webhook payload includes: -

startedAt

,

finishedAt

-

durationInSeconds

,

durationFormatted

-

success

flag -

errorType

,

errorOrigin

(for failures) Example for a failed sync webhook:[[Webhook payloads](https://docs.airbyte.com/platform/cloud/managing-airbyte-cloud/manage-airbyte-cloud-notifications#set-up-webhook-notifications)]

json    {      "data": {        "workspace": {...},        "connection": {...},        "jobId": 9988,        "startedAt": "2024-01-01T00:00:00Z",        "finishedAt": "2024-01-01T01:00:00Z",        "success": false,        "durationInSeconds": 3600,        "durationFormatted": "1 hours 0 min"      }    }

You can use these fields in your webhook receiver to implement your own logic, for example: - Compare

durationInSeconds

to historical averages and alert if it’s unusually high. - Alert when multiple failed syncs occur in a row for the same connection (in addition to the built‑in “repeated failures” events). 3. Platform‑level safeguards against hanging jobs Airbyte has internal mechanisms that detect when a sync is “stuck” and will eventually fail it, which in turn can trigger your failure notifications: - Source heartbeats: If the source stops emitting

RECORD

or

STATE

messages for longer than

heartbeat-max-seconds-between-messages

(default 3 hours), the sync is failed with “The source is unresponsive.”[[Heartbeats](https://docs.airbyte.com/platform/understanding-airbyte/heartbeats#technical-details)] - Destination timeout: If a call to the destination exceeds

destination-timeout-max-seconds

(default 24 hours), the sync is failed with “The destination is unresponsive.”[[Heartbeats](https://docs.airbyte.com/platform/understanding-airbyte/heartbeats#technical-details)] Both are configurable in

flags.yaml

and, when they cause a failure, you can be notified via the same notification system. 4. Global job timeout (older versions) In older forum posts, support mentions

SYNC_JOB_MAX_TIMEOUT_DAYS

(minimum 1 day) to forcefully time out very long‑running jobs:[[Sync timeout env var](https://discuss.airbyte.io/t/709)]

env    SYNC_JOB_MAX_TIMEOUT_DAYS=3

The knowledge sources don’t document a more granular hours‑level timeout, only this days‑level setting and the heartbeat/timeout mechanisms above. --- So while there isn’t a built‑in “alert when this specific sync exceeds X minutes” toggle in the docs, you can: - Enable Failed syncs and repeated‑failure notifications (email or webhook). - Use the webhook payload to implement your own “slower than usual” or “appears stalled” logic in whatever monitoring/alerting system you use.

How does powered by airbyte work in terms of pricing? we self-host and are wondering if we could do the same with a headless stack.