Up: Payment System Observability See also: Detecting Stripe Incidents This page is a use-case implementation of the mechanics described in Multi-PSP Payment Risk Observability
Cross-PSP Failures
Definition
Cross-PSP Detection identifies systemic payment failures that affect multiple processors simultaneously. It differentiates "My Processor is down" from "The Card Network is down."
Why it matters
Routing logic. If Stripe is down, you route to Adyen. If Visa is down, routing to Adyen is useless (and costs fees). Identifying the root scope prevents wasted retries and allows for accurate status page communication.
Signals to monitor
- Global Error Sync: Are
500errors spiking on Stripe AND PayPal at the exact same second? - Issuer Correlation: Are all processors declining
Chasecards specifically? - Region Correlation: Are all processors failing for
UKIP addresses?
Breakdown modes
- 3DS Outage: The 3D Secure directory server goes down; all processors fail auths requiring challenge.
- AWS East Outage: Underlying cloud infra impacting multiple providers.
- Bin Attack: A global fraud attack hitting all your gateways at once.
Where observability fits
- Triangulation: "Adyen is healthy. Stripe is failing. conclusion: Stripe Incident."
- Status Page Independence: Knowing about an outage 15 minutes before the provider updates their status page.
- Smart Routing: Auomatically disabling a specific path based on error rate.
Note: observability does not override processor or network controls; it provides operational clarity to navigate them.
FAQ
How common are Network outages?
Rare for the core network (Visa/MC). More common for edge services (3DS, Risk APIs).
Should I retry on a different PSP?
Only if the error is "Gateway Error" or "Timeout." Never retry "Insufficient Funds" or "Do Not Honor" across PSPs (it looks like velocity fraud).