Microservices Resilience and Fault Tolerance 5

Question 1

What is the primary risk of a 'retry storm' across many client instances?

Accepted Answer

Synchronized retries that suddenly multiply load and prolong or trigger an outage

Answer

A retry storm happens when many clients retry in unison, spiking load on a recovering service and worsening the situation.

Question 2

Which metric is most useful for tuning a circuit breaker's failure threshold?

Accepted Answer

The error or failure rate of calls within a rolling window

Answer

Circuit breakers typically trip based on the proportion of failed calls in a recent sliding window.

Question 3

What is the benefit of caching responses as a fallback during a downstream outage?

Accepted Answer

It allows serving slightly stale but usable data instead of an error

Answer

Serving cached (stale) data during an outage keeps functionality available, trading freshness for availability.

Question 4

In distributed systems, why does the timeout budget need to be coordinated across a call chain?

Accepted Answer

So an upstream timeout isn't shorter than the sum of downstream timeouts, causing wasted work

Answer

Timeout budgets must decrease down the chain so callers don't give up while downstream calls still run and waste resources.

Question 5

What does the term 'blast radius' mean in resilience design?

Accepted Answer

The scope of impact a single failure can have on the rest of the system

Answer

Blast radius describes how far a failure can spread; isolation patterns aim to keep it small.

Microservices Practice Test