Live status, reported honestly.
If a customer's line stops because of us, it appears here — with the real timeline, not a sanitised version published a week later.
Last updated just now · times shown in IST (UTC+5:30)
Service by service.
Each bar represents one day over the last 60 days. Hover a bar to see its date. Amber indicates degraded performance; red indicates an outage.
Incident reports.
Every incident gets a public write-up with a root cause and the change we made.
There are currently no open incidents. The most recent resolved incident is listed below.
Duration: 41 minutes · Impact: Analytics dashboards slow, serialization unaffected
Root cause. An index rebuild on the scan-events table coincided with an unusually large nightly batch from a high-volume customer, saturating IO on the analytics replica.
What we changed. Index rebuilds are now rate-limited and excluded from peak ingestion windows. We also added IO saturation alerting with a five-minute warning threshold, and moved the two largest tenants to dedicated analytics replicas.
Serialization, verification and consumer pages were not affected. No customer data was lost.
Duration: 1 hour 12 minutes · Impact: Delayed webhook delivery to ERP integrations
Root cause. A downstream ERP endpoint at a single customer began responding slowly, exhausting a shared delivery worker pool. The backpressure was not isolated per tenant.
What we changed. Webhook delivery pools are now isolated per tenant with circuit breakers, so a slow endpoint at one customer cannot degrade delivery for another. We also exposed per-integration delivery lag in the console.
Scan ingestion and verification were unaffected throughout.
Get incident notifications
Subscribe to status updates by email. You will only receive messages when something changes — no monthly digest, no marketing.