Redis哨兵远程连接问题求助：集群环境部署遇阻

Hey there, let's work through why your Redis Sentinel can't establish a remote connection to your master node — I’ve been stuck with this exact problem before, so here are the key checks and fixes to try:

1. Verify Master Node Configuration #

First, let’s rule out issues with your master Redis instance:

• Bind Address & Protected Mode: You mentioned your master binds to 127.0.0.1 and a 192.x.x.x IP. Double-check that protected-mode no is set in your redis.conf — Redis’s default protected mode blocks external connections even if you’ve bound to a remote IP. If you don’t want to disable protected mode entirely, you can add your Sentinel nodes’ IPs to an ACL list or set a requirepass (and make sure Sentinel uses it, more on that later).
• Confirm Correct Bind IP: Ensure the 192.x.x.x IP in your config matches the master node’s actual public/network-facing IP. Run ip addr (Linux) or ipconfig (Windows) on the master to verify the correct interface IP.

2. Check Network & Firewall Accessibility #

This is the most common culprit for remote connection failures:

• Test Basic Connectivity: On your Sentinel node, run redis-cli -h <master-ip> -p 6379 to see if you can manually connect to the master. If this fails, try telnet <master-ip> 6379 or nc -zv <master-ip> 6379 to confirm if the port is reachable.
• Firewall/Security Group Rules: Make sure the master node’s firewall allows incoming traffic on port 6379 (Redis default). If you’re using cloud servers (AWS/Azure/GCP), check your security groups to ensure the Sentinel nodes’ IPs are allowed to access port 6379. On Linux, use ufw allow 6379 or iptables -A INPUT -p tcp --dport 6379 -s <sentinel-ip> -j ACCEPT to open the port.

3. Validate Sentinel Configuration #

Even if the master is reachable, misconfigured Sentinel settings will break the connection:

• Correct Master Monitor Directive: In your Sentinel config, ensure you’re pointing to the right master IP and port:

  sentinel monitor mymaster <master-ip> 6379 2
  

  The 2 here is the quorum (number of Sentinels needed to agree on a failover) — this is correct for 3 Sentinel nodes.
• Authentication Setup: If your master has a requirepass set in its redis.conf, you must add the corresponding auth pass to Sentinel’s config:

  sentinel auth-pass mymaster <your-master-password>
  

• Sentinel Bind Setting: Don’t let Sentinel only bind to 127.0.0.1 — set bind 0.0.0.0 or the Sentinel node’s network-facing IP in sentinel.conf so it can communicate with remote nodes.

4. Dig Into Logs for Specific Errors #

Logs will tell you exactly what’s going wrong:

• Check your master Redis log (default path: /var/log/redis/redis-server.log or the path set in logfile in redis.conf) for lines like protected mode enabled, connection from <sentinel-ip> refused or NOAUTH Authentication required.
• Check Sentinel’s log (default: /var/log/redis/redis-sentinel.log) for entries like Can't connect to master or Authentication error.

Quick Test Flow #

1. Fix any connectivity issues with manual redis-cli access to the master first.
2. Once manual access works, restart your Sentinel service and check its logs again.
3. If Sentinel still can’t connect, double-check all config directives and firewall rules one more time.

内容的提问来源于stack exchange，提问作者John