Adding and removing nodes in Redis

Introduction

Adding and removing nodes helps you grow or shrink your Redis cluster to handle more data or save resources.

When your Redis cluster needs more capacity to store data.

When you want to improve performance by distributing load.

When you want to remove a faulty or old node from the cluster.

When you want to save costs by reducing the number of nodes.

When you are scaling your application up or down.

Syntax

Redis

redis-cli --cluster add-node <new-node-ip>:<port> <existing-node-ip>:<port>
redis-cli --cluster del-node <existing-node-ip>:<port> <node-id>

Use : to specify the new node you want to add.

Use : to specify a node already in the cluster.

Node ID is a unique identifier for each node in the cluster.

Examples

Adds a new node at 192.168.1.10:7003 to the cluster that already has a node at 192.168.1.10:7000.

Redis

redis-cli --cluster add-node 192.168.1.10:7003 192.168.1.10:7000

Removes the node with ID 07c37dfeb2352e03a1b7f4a1a3e8f9c8b7d6a1f2 from the cluster using the node at 192.168.1.10:7000 as reference.

Redis

redis-cli --cluster del-node 192.168.1.10:7000 07c37dfeb2352e03a1b7f4a1a3e8f9c8b7d6a1f2

Sample Program

First, this command adds a new node at 127.0.0.1:7003 to the cluster that already has a node at 127.0.0.1:7000. Then, it removes a node by its ID from the cluster.

Redis

redis-cli --cluster add-node 127.0.0.1:7003 127.0.0.1:7000
redis-cli --cluster del-node 127.0.0.1:7000 07c37dfeb2352e03a1b7f4a1a3e8f9c8b7d6a1f2

Important Notes

Always use a node already in the cluster as a reference when adding or removing nodes.

Removing a node will migrate its data to other nodes if needed.

Make sure the new node is running Redis and reachable before adding.

Summary

Adding nodes helps your cluster grow and handle more data.

Removing nodes helps you shrink the cluster or remove bad nodes.

Use redis-cli with --cluster add-node and --cluster del-node commands.

Practice

1. What is the primary purpose of adding nodes to a Redis cluster?

easy

A. To increase the cluster's capacity and distribute data

B. To delete all data from the cluster

C. To backup the cluster data

D. To change the Redis version on all nodes

Solution

1. Step 1: Understand what adding nodes means

   Adding nodes means increasing the number of servers in the cluster to share the data load.

2. Step 2: Identify the effect of adding nodes

   More nodes mean more capacity and better data distribution across the cluster.

3. Final Answer:

   To increase the cluster's capacity and distribute data -> Option A

4. Quick Check:

   Adding nodes = More capacity and distribution [OK]

Hint: Adding nodes grows cluster capacity and data spread [OK]

Common Mistakes:

• Thinking adding nodes deletes data
• Confusing adding nodes with backup
• Assuming adding nodes upgrades Redis version

2. Which of the following is the correct syntax to add a node to a Redis cluster using redis-cli?

easy

A. redis-cli --cluster add-node :<port> :<port>

B. redis-cli --cluster remove-node :<port>

C. redis-cli --cluster create-node :<port>

D. redis-cli --cluster add :<port>

Solution

1. Step 1: Recall the command to add a node

   The correct command uses --cluster add-node followed by the new node and an existing node in the cluster.

2. Step 2: Match the syntax

   redis-cli --cluster add-node :<port> :<port> matches the correct syntax: redis-cli --cluster add-node new-node existing-node.

3. Final Answer:

   redis-cli --cluster add-node <new-node-ip>:<port> <existing-node-ip>:<port> -> Option A

4. Quick Check:

   Correct add-node syntax = redis-cli --cluster add-node :<port> :<port> [OK]

Hint: Add node command uses --cluster add-node with new and existing nodes [OK]

Common Mistakes:

• Using remove-node instead of add-node
• Using create-node which does not exist
• Omitting existing node IP in command

3. Given the command redis-cli --cluster del-node 192.168.1.5:7000 07c37dfeb2352e03a5a1e1f3a2f8b9a1f2e3d4c5, what will happen?

medium

A. A new node will be added at 192.168.1.5:7000

B. The node at 192.168.1.5:7000 will be removed from the cluster

C. The cluster will be created with the given node ID

D. The command will fail due to wrong syntax

Solution

1. Step 1: Understand the del-node command

   The command redis-cli --cluster del-node removes a node from the cluster using its IP and node ID.

2. Step 2: Analyze the given command

   The command specifies the node IP and the node ID, so it will remove that node from the cluster.

3. Final Answer:

   The node at 192.168.1.5:7000 will be removed from the cluster -> Option B

4. Quick Check:

   del-node removes specified node [OK]

Hint: del-node removes node by IP and node ID [OK]

Common Mistakes:

• Confusing del-node with add-node
• Thinking it adds a node
• Assuming command syntax is incorrect

4. You run redis-cli --cluster add-node 192.168.1.10:7003 192.168.1.11:7000 but get an error. What is the most likely cause?

medium

A. The new node is already part of the cluster

B. The existing node IP is incorrect

C. The command should be --cluster del-node instead

D. Redis server is not running on the new node

Solution

1. Step 1: Check prerequisites for adding a node

   The new node must be running Redis and reachable for the add-node command to succeed.

2. Step 2: Identify common causes of add-node failure

   If Redis is not running on the new node, the command will fail because it cannot connect.

3. Final Answer:

   Redis server is not running on the new node -> Option D

4. Quick Check:

   New node must run Redis to add [OK]

Hint: New node must be running Redis before adding [OK]

Common Mistakes:

• Assuming IP typo without checking server status
• Confusing add-node with del-node command
• Ignoring that Redis must run on new node

5. You want to remove a node from a Redis cluster but keep its data safe. Which sequence of commands should you use?

hard

A. Use redis-cli --cluster del-node then backup data manually

B. Directly stop the Redis server on the node and remove it

C. Use redis-cli --cluster reshard to move slots, then redis-cli --cluster del-node

D. Use redis-cli --cluster add-node to replace the node

Solution

1. Step 1: Understand data safety when removing nodes

   Removing a node without moving its data causes data loss.

2. Step 2: Use reshard to move data slots

   Resharding moves the data slots from the node to be removed to other nodes safely.

3. Step 3: Remove the node after resharding

   After data is moved, del-node safely removes the node without data loss.

4. Final Answer:

   Use redis-cli --cluster reshard to move slots, then redis-cli --cluster del-node -> Option C

5. Quick Check:

   Reshard then del-node = safe removal [OK]

Hint: Reshard slots before removing node to keep data safe [OK]

Common Mistakes:

• Removing node without moving data first
• Stopping server directly causing data loss
• Confusing add-node with removal process