[{"data":1,"prerenderedAt":10},["ShallowReactive",2],{"public:public\u002Fcontent\u002Fdocs\u002Fwebhooks":3},{"slug":4,"title":5,"description":6,"draft":7,"updated":8,"html":9},"webhooks","Webhooks and incident tools","Signed webhooks, Microsoft Teams, PagerDuty and Opsgenie alerts for restore test results, with retries and a delivery log.",false,null,"\u003Ch1>Webhooks, Microsoft Teams, PagerDuty and Opsgenie\u003C\u002Fh1>\n\u003Cp>Add channels under \u003Cstrong>Alerts\u003C\u002Fstrong>. Every channel carries the same facts as e-mail and Slack alerts: the target name, the status and reason, timings, and the IDs of the checks that did not pass. Never check messages, observed values or anything from your database.\u003C\u002Fp>\n\u003Ch2>Signed webhook\u003C\u002Fh2>\n\u003Cp>Verestore sends a \u003Ccode>POST\u003C\u002Fcode> with a JSON body to a public \u003Ccode>https:\u002F\u002F\u003C\u002Fcode> URL of your choice for each judged run (or only for problems):\u003C\u002Fp>\n\u003Cpre>\u003Ccode class=\"language-json\">{\n  "type": "run.judged",\n  "id": "whd_01JBX…",\n  "created_at": "2026-11-05T02:14:07.000Z",\n  "data": {\n    "run_id": "run_01JBX…", "org_id": "org_01JBX…",\n    "target": { "id": "tgt_01JBX…", "name": "orders-db" },\n    "status": "failed", "status_reason": null, "judged_at": "2026-11-05T02:14:07.000Z",\n    "restore_seconds": 184.6, "checks_not_passing": ["row_counts"],\n    "url": "https:\u002F\u002Fapp.verestore.dev\u002Fo\u002Forg_01JBX…\u002Fruns\u002Frun_01JBX…"\n  }\n}\n\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>Headers: \u003Ccode>X-Verestore-Event\u003C\u002Fcode> (\u003Ccode>run.judged\u003C\u002Fcode> or \u003Ccode>test\u003C\u002Fcode>), \u003Ccode>X-Verestore-Delivery\u003C\u002Fcode> (the same on every retry: use it to deduplicate) and \u003Ccode>X-Verestore-Signature: t=<unix seconds>,v1=<hex>\u003C\u002Fcode>.\u003C\u002Fp>\n\u003Cp>The signing secret (\u003Ccode>whsec_…\u003C\u002Fcode>) is shown once, when you add the channel. To verify a delivery, compute HMAC-SHA256 over \u003Ccode><t>.<raw body>\u003C\u002Fcode> with the secret, compare it in constant time with \u003Ccode>v1\u003C\u002Fcode>, and refuse timestamps more than 5 minutes old:\u003C\u002Fp>\n\u003Cpre>\u003Ccode class=\"language-js\">import { createHmac, timingSafeEqual } from 'node:crypto'\n\nfunction verify(secret, rawBody, header, now = Date.now()) {\n  const m = \u002F^t=(\\d+),v1=([0-9a-f]{64})$\u002F.exec(header ?? '')\n  if (!m || Math.abs(now \u002F 1000 - Number(m[1])) > 300) return false\n  const want = createHmac('sha256', secret).update(`${m[1]}.${rawBody}`).digest()\n  return timingSafeEqual(want, Buffer.from(m[2], 'hex'))\n}\n\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>Answer with any \u003Ccode>2xx\u003C\u002Fcode> status within 10 seconds. Redirects are not followed, and the URL must resolve to a public address: this is checked when you add the channel and again on every delivery.\u003C\u002Fp>\n\u003Ch2>Microsoft Teams\u003C\u002Fh2>\n\u003Cp>Create a \u003Cstrong>Workflows\u003C\u002Fstrong> flow "Post to a channel when a webhook request is received" (or an incoming webhook connector) and paste its URL. Verestore posts an Adaptive Card with the status, timings, the checks that did not pass and an \u003Cstrong>Open run\u003C\u002Fstrong> button. Only Microsoft webhook hosts are accepted.\u003C\u002Fp>\n\u003Ch2>PagerDuty\u003C\u002Fh2>\n\u003Cp>Add an \u003Cstrong>Events API v2\u003C\u002Fstrong> integration to a service and paste its integration key; choose the EU region if your account is hosted there. A warned, failed or inconclusive run triggers an incident with deduplication key \u003Ccode>verestore:<target id>\u003C\u002Fcode> (so repeated failures of one target update one incident), and the next passing run of that target resolves it.\u003C\u002Fp>\n\u003Ch2>Opsgenie\u003C\u002Fh2>\n\u003Cp>Create an \u003Cstrong>API\u003C\u002Fstrong> integration and paste its key; choose the EU region if your account is hosted there. A problem run creates an alert with alias \u003Ccode>verestore:<target id>\u003C\u002Fcode> (P2 for failed, P3 otherwise); the next passing run of that target closes it.\u003C\u002Fp>\n\u003Ch2>Retries and the delivery log\u003C\u002Fh2>\n\u003Cp>Webhook, Teams, PagerDuty and Opsgenie deliveries are retried on network errors, timeouts, \u003Ccode>408\u003C\u002Fcode>, \u003Ccode>429\u003C\u002Fcode> and \u003Ccode>5xx\u003C\u002Fcode>: after 1 minute, 5 minutes, 15 minutes, 1 hour, 3 hours, 6 hours and 12 hours, for at most 24 hours from the first attempt. Other \u003Ccode>4xx\u003C\u002Fcode> answers are final. The \u003Cstrong>Delivery log\u003C\u002Fstrong> on the Alerts page shows each delivery's status, HTTP status, number of attempts and next attempt. \u003Cstrong>Send test\u003C\u002Fstrong> delivers a test message once, without retries.\u003C\u002Fp>\n",1794290783592]