Concept questionMedium~7 min

Interview question

Redis is unavailable during peak traffic. How should the API behave?

Handle Redis outage with fallback and degraded mode

Explains how an API behaves when Redis is unavailable and which cache usages can safely fall back.

1Say itAnswer out loud first

2CompareCompare with Core answer

3RefineRefine with Stronger and Advanced

TL;DR

If Redis is only used as a read-through cache, the API can often treat failures as cache misses and read from the database. But it must protect the database because a Redis outage can suddenly send much more traffic to the source. If Redis holds session state, rate-limit counters, locks, or workflow state, the fallback behavior needs a more explicit design.

Full answer below

Practice focus

Dependency failure, fallback reads, degraded behavior, source protection, circuit breakers, and user-facing impact.

Answer first

Answer path

Core answer first, then checkpoints toward a stronger answer.

Start
Core answer
Start here. This is the direct spoken answer to practice first.
Why this question matters
A Redis outage reveals whether each usage is optional acceleration, security control, coordination, or undeclared state. Fallback and fail-closed decisions must be defined per usage before peak traffic reaches the source system.
If Redis is only used as a read-through cache, the API can often treat failures as cache misses and read from the database. But it must protect the database because a Redis outage can suddenly send much more traffic to the source. If Redis holds session state, rate-limit counters, locks, or workflow state, the fallback behavior needs a more explicit design.
Checkpoint
Stronger answer
Optional upgrade with examples, trade-offs, and sharper phrasing.
I would classify each Redis usage. For product data cache, fallback to source with rate limits or degraded pages may be acceptable. For sessions, users may need to reauthenticate or the app may need a durable session store. For rate limiting, the safe fallback might be fail-closed for risky endpoints or a local temporary limiter for low-risk traffic. For locks and coordination, continuing blindly can create duplicate work.
Finish
Advanced answer
Optional deeper angle for production judgment and nuance.
In production every Redis usage has a failure classification and fallback budget. Optional reads use short timeouts, circuit breaking, bounded source fallback, and load shedding so a cache outage cannot overwhelm SQL. Sessions, rate limits, and coordination choose fail-open or fail-closed from security risk rather than convenience. Recovery is gradual because reconnecting every instance and refilling every key can create another spike. Dashboards connect Redis errors, fallback rate, source saturation, rejected traffic, and affected features to a runbook for safe degradation and restoration.

Review signals

Use these to catch weak phrasing, shallow reasoning, and risky claims.

Common mistakesCandidateCommon mistakesMistakes that weaken the answer or signal shallow understanding.Open details

Details

Assuming Redis can fail without any user impact.
Falling back to the database with no source protection.
Treating sessions, locks, counters, and read cache the same way.
Retrying Redis aggressively during an outage.

Red flagsCandidateRed flagsSignals that an interviewer may treat as risky.Open details

Details

Cannot classify Redis usage by criticality.
No degraded-mode plan.
No timeout or circuit-breaker boundary.
No monitoring for fallback pressure.

Follow-up prompts

Quick interviewer prompts inside this drill.

Follow-up questionsCandidateFollow-up questionsLikely next questions after the main answer.Open details

Details

Treat these as the next interviewer turn. Answer one out loud, then save the gap in your practice notes.

Which Redis uses can fail open?
Which should fail closed?
How would you protect SQL during a Redis outage?

Continue practice

Move to the linked follow-up, next path step, prerequisite, or deeper variant.

1 linked

Try a variant

Answer a similar question in a different context.

Dependency Timeout Incident

Concept questionMedium~7 min

Interview question

Redis is unavailable during peak traffic. How should the API behave?

Handle Redis outage with fallback and degraded mode

Explains how an API behaves when Redis is unavailable and which cache usages can safely fall back.

1Say itAnswer out loud first

2CompareCompare with Core answer

3RefineRefine with Stronger and Advanced

TL;DR

Full answer below

Practice focus

Dependency failure, fallback reads, degraded behavior, source protection, circuit breakers, and user-facing impact.

Answer first

Answer path

Core answer first, then checkpoints toward a stronger answer.

Start
Core answer
Start here. This is the direct spoken answer to practice first.
Why this question matters
A Redis outage reveals whether each usage is optional acceleration, security control, coordination, or undeclared state. Fallback and fail-closed decisions must be defined per usage before peak traffic reaches the source system.
If Redis is only used as a read-through cache, the API can often treat failures as cache misses and read from the database. But it must protect the database because a Redis outage can suddenly send much more traffic to the source. If Redis holds session state, rate-limit counters, locks, or workflow state, the fallback behavior needs a more explicit design.
Checkpoint
Stronger answer
Optional upgrade with examples, trade-offs, and sharper phrasing.
I would classify each Redis usage. For product data cache, fallback to source with rate limits or degraded pages may be acceptable. For sessions, users may need to reauthenticate or the app may need a durable session store. For rate limiting, the safe fallback might be fail-closed for risky endpoints or a local temporary limiter for low-risk traffic. For locks and coordination, continuing blindly can create duplicate work.
Finish
Advanced answer
Optional deeper angle for production judgment and nuance.
In production every Redis usage has a failure classification and fallback budget. Optional reads use short timeouts, circuit breaking, bounded source fallback, and load shedding so a cache outage cannot overwhelm SQL. Sessions, rate limits, and coordination choose fail-open or fail-closed from security risk rather than convenience. Recovery is gradual because reconnecting every instance and refilling every key can create another spike. Dashboards connect Redis errors, fallback rate, source saturation, rejected traffic, and affected features to a runbook for safe degradation and restoration.

Review signals

Use these to catch weak phrasing, shallow reasoning, and risky claims.

Common mistakesCandidateCommon mistakesMistakes that weaken the answer or signal shallow understanding.Open details

Details

Assuming Redis can fail without any user impact.
Falling back to the database with no source protection.
Treating sessions, locks, counters, and read cache the same way.
Retrying Redis aggressively during an outage.

Red flagsCandidateRed flagsSignals that an interviewer may treat as risky.Open details

Details

Cannot classify Redis usage by criticality.
No degraded-mode plan.
No timeout or circuit-breaker boundary.
No monitoring for fallback pressure.

Follow-up prompts

Quick interviewer prompts inside this drill.

Follow-up questionsCandidateFollow-up questionsLikely next questions after the main answer.Open details

Details

Treat these as the next interviewer turn. Answer one out loud, then save the gap in your practice notes.

Which Redis uses can fail open?
Which should fail closed?
How would you protect SQL during a Redis outage?

Continue practice

Move to the linked follow-up, next path step, prerequisite, or deeper variant.

1 linked

Try a variant

Answer a similar question in a different context.

Dependency Timeout Incident

Redis is unavailable during peak traffic. How should the API behave?

Practice focus

Answer path

Why this question matters

Review signals

Follow-up prompts

Continue practice

Try a variant

Redis is unavailable during peak traffic. How should the API behave?

Practice focus

Answer path

Why this question matters

Review signals

Follow-up prompts

Continue practice

Try a variant