Multiple Sentinel instances in Redis - Practice Problems & Coding Challenges

Challenge - 5 Problems

🧠 Conceptual

intermediate

Purpose of Multiple Sentinel Instances

Why is it recommended to run multiple Redis Sentinel instances monitoring the same master?

A To enable Redis to store more data by distributing it across Sentinels

B To increase the chance of automatic failover by achieving a quorum among Sentinels

C To allow clients to connect to multiple masters simultaneously

D To reduce the memory usage on the Redis master server

💡 Hint

Think about how Sentinel decides to promote a new master.

Multiple Sentinel instances provide fault tolerance by voting to agree on failover decisions. This quorum prevents false failovers.

❓ query_result

intermediate

Sentinel Master Status Query Result

Given three Sentinel instances monitoring a Redis master, what will the SENTINEL masters command return?

Redis

SENTINEL masters

A An error because multiple Sentinels cannot run simultaneously

B A list of all connected clients to the master

C A simple string with the master name only

D A list of dictionaries with details about each monitored master, including name, IP, port, and flags

💡 Hint

Consider what information Sentinel provides about monitored masters.

The SENTINEL masters command returns detailed info about each master monitored by Sentinel instances.

📝 Syntax

advanced

Correct Sentinel Configuration for Multiple Instances

Which configuration snippet correctly sets up three Sentinel instances monitoring the same Redis master?

A

sentinel monitor mymaster 192.168.1.100 6379 2
sentinel down-after-milliseconds mymaster 5000
sentinel failover-timeout mymaster 10000

B

00001 retsamym tuoemit-revoliaf lenitnes
0005 retsamym sdnocesillim-retfa-nwod lenitnes
2 9736 001.1.861.291 retsamym rotinom lenitnes

C

sentinel monitor mymaster 192.168.1.100 6379 1
sentinel down-after-milliseconds mymaster 5000
sentinel failover-timeout mymaster 10000

D

sentinel monitor mymaster 192.168.1.100 6379 0
sentinel down-after-milliseconds mymaster 5000
sentinel failover-timeout mymaster 10000

💡 Hint

The quorum number must be the minimum number of Sentinels that agree to failover.

The quorum value in sentinel monitor must be at least 2 when running three Sentinels to ensure majority agreement.

❓ optimization

advanced

Optimizing Sentinel Failover Decision Time

You want to reduce the time Sentinel takes to detect a master failure and start failover. Which configuration change helps achieve this without risking false failovers?

A Increase both <code>down-after-milliseconds</code> and <code>failover-timeout</code>

B Increase <code>down-after-milliseconds</code> and decrease <code>failover-timeout</code>

C Decrease both <code>down-after-milliseconds</code> and <code>failover-timeout</code> carefully

D Decrease <code>down-after-milliseconds</code> and increase <code>failover-timeout</code>

💡 Hint

Think about how quickly Sentinel detects failure and how long it waits before failover.

Reducing down-after-milliseconds makes Sentinel detect failure faster; reducing failover-timeout speeds failover start. Both must be balanced to avoid false failovers.

🔧 Debug

expert

Diagnosing Sentinel Failover Failure

You have three Sentinel instances monitoring a Redis master with quorum set to 2. After the master crashes, failover does not occur. What is the most likely cause?

A At least two Sentinels did not detect the master as down, so quorum was not reached

B The master was not configured with <code>replicaof</code> pointing to Sentinels

C The Sentinels were running on the same server, causing network partition

D The <code>failover-timeout</code> was set to zero, disabling failover

💡 Hint

Consider how Sentinel decides to start failover based on quorum.

Failover requires a quorum of Sentinels to agree the master is down. If fewer than quorum detect failure, failover won't start.