consul watch, the server is offline and will not receive any notification

[noovertime7]

What happened:

使用consul作为注册中心，当服务端离线时并不会收到任何通知，无法触发离线的回调

What you expected to happen:

服务端离线使用watch收到通知

How to reproduce it (as minimally and precisely as possible):

Anything else we need to know?:

使用etcd是可以的

Environment:

• Kratos version (use kratos -v):kratos version v2.7.3
• Go version (use go version): go version go1.21.2 darwin/arm64
• OS (e.g: cat /etc/os-release): mac book
• Others:

[dosubot]

你好，@noovertime7！我是 Dosu，一个可以帮助你解决问题、回答问题以及成为贡献者的机器人。我在这里帮助你，同时等待人类维护者的回复。我会立即查看你的问题，并尽快给你一个完整的答案。

[kratos-ci-bot]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

Hello, @noovertime7! I'm Dosu, a bot that helps you solve problems, answer questions, and become a contributor. I'm here to help you while I wait to hear back from the human maintainers. I will review your question immediately and give you a complete answer as soon as possible.

[noovertime7]

只有服务端上线时会收到消息，离线不会

[kratos-ci-bot]

Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑‍🤝‍🧑👫🧑🏿‍🤝‍🧑🏻👩🏾‍🤝‍👨🏿👬🏿

---

The message will only be received when the server is online, not when offline.

[dosubot]

To address the issue where Consul watch does not trigger notifications when the server goes offline, unlike with etcd, you might need to implement additional logic in your Kratos project. The current implementation in Kratos for Consul watches, as seen in contrib/registry/consul/watcher.go, waits for an event or the context to be done without specifically handling server offline events.

Here's a conceptual approach to enhance the watch functionality to detect when the Consul server goes offline:

1. Implement Health Checks: Create health checks for the Consul server. When the server is detected as offline, trigger an event on the w.event channel.

2. Extend the Watch Mechanism: Modify the existing watch mechanism to periodically verify the connectivity to the Consul server. On detecting a disconnection, trigger an event on the w.event channel.

3. Utilize Consul Events: Leverage Consul's event system to detect server offline events and integrate them with the Kratos service discovery watch mechanism.

For example, you could implement a health check that triggers an event when the server goes offline as follows:

func (w *watcher) monitorConsulServer() {
    ticker := time.NewTicker(10 * time.Second)
    defer ticker.Stop()

    for {
        select {
        case <-w.ctx.Done():
            return
        case <-ticker.C:
            if !isConsulServerAlive() {
                w.event <- struct{}{}
            }
        }
    }
}

func isConsulServerAlive() bool {
    // Logic to check if the Consul server is alive, e.g., a ping or health check
    return true // Placeholder return value
}

This function should be called as a goroutine, possibly during the watcher's initialization. Adjust this example based on your specific requirements and environment. For more details on the watcher implementation, refer to the Kratos repository, specifically the contrib/registry/consul/watcher.go file.

Details

Sources

• contrib/registry/consul/watcher.go
• contrib/config/etcd/watcher.go
• contrib/registry/etcd/watcher.go
• contrib/registry/consul/service.go
• contrib/registry/consul/registry.go
• contrib/config/consul/watcher.go

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.

[shenqidebaozi]

I'm not sure what you're referring to. I've been using this feature all along. Please provide a complete explanation or reproduce an example