Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Task] hertzbeat metrics collector cluster design #1003

Closed
tomsun28 opened this issue May 31, 2023 · 12 comments · Fixed by #1101
Closed

[Task] hertzbeat metrics collector cluster design #1003

tomsun28 opened this issue May 31, 2023 · 12 comments · Fixed by #1101
Labels
doc Improvements or additions to documentation enhancement New feature or request good first issue Good for newcomers new feature task

Comments

@tomsun28
Copy link
Contributor

Description

why collector cluster?

  • cluster means more powerful performance and can support monitoring more
  • unified monitoring between different isolated networks, one hertzbeat can monitoring multi isolated networks resource.
  • avoid single point of failure

desgin discuss

image

Task List

No response

@tomsun28 tomsun28 added doc Improvements or additions to documentation new feature labels May 31, 2023
@Ceilzcx
Copy link
Contributor

Ceilzcx commented May 31, 2023

讨论存疑:

  1. 采集后数据的存储位置?
    主hz(collector每次采集完需要将数据同步到主hz,需要处理各种丢包等问题,带宽使用高)
    collector模块自己存储(无法实现内网的collector,通过外网查询)
  2. hz如何配置collector模块的地址?
    配置文件
    数据库
  3. 心跳机制,collector向hz发送心跳还是hz向collector发送心跳?
  4. hz如何响应collector模块的上线,注销,连接失败,重连接等问题?

@hertzbeat
Copy link
Contributor

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


Discussion in doubt:

  1. Where is the collected data stored?
    Main hz (the collector needs to synchronize the data to the main hz after each collection, and needs to deal with various packet loss and other problems, and the bandwidth usage is high)
    The collector module stores itself (the collector of the intranet cannot be realized, and it can be queried through the external network)
  2. How does hz configure the address of the collector module?
    configuration file
    database
  3. Heartbeat mechanism, does the collector send a heartbeat to hz or does hz send a heartbeat to the collector?
  4. How does hz respond to problems such as online, logout, connection failure, and reconnection of the collector module?

@zqr10159
Copy link
Member

在监控量不大的情况下,是否保留原来的模式,以简化部署难度?当初就是看着hz不用额外部署agent才用的

@hertzbeat
Copy link
Contributor

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


In the case of a small amount of monitoring, should the original mode be retained to simplify the difficulty of deployment? At the beginning, it was only used when looking at hz without additional deployment of agent

@tomsun28
Copy link
Contributor Author

在监控量不大的情况下,是否保留原来的模式,以简化部署难度?当初就是看着hz不用额外部署agent才用的

是的,默认还是会单体 all in one. 设计时也会尽量减少外部组件依赖

@hertzbeat
Copy link
Contributor

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


In the case of a small amount of monitoring, is the original mode retained to simplify the difficulty of deployment? At the beginning, it was only used when looking at hz without additional deployment of agent

Yes, the default is still all in one. The design will also minimize external component dependencies

@l646505418
Copy link
Contributor

在监控量不大的情况下,是否保留原来的模式,以简化部署难度?当初就是看着hz不用额外部署agent才用的

是的,默认还是会单体 all in one. 设计时也会尽量减少外部组件依赖

如果在主 hz中提供webhook來接收其他collector的信息会不会减少对kafka的依赖

@hertzbeat
Copy link
Contributor

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


In the case of a small amount of monitoring, is the original mode retained to simplify the difficulty of deployment? At the beginning, it was only used when looking at hz without additional deployment of agent

Yes, the default is still all in one. The design will also minimize external component dependencies

If webhook is provided in the main hz to receive information from other collectors, will it reduce the dependence on Kafka?

@tomsun28
Copy link
Contributor Author

tomsun28 commented Jun 2, 2023

在监控量不大的情况下,是否保留原来的模式,以简化部署难度?当初就是看着hz不用额外部署agent才用的

是的,默认还是会单体 all in one. 设计时也会尽量减少外部组件依赖

如果在主 hz中提供webhook來接收其他collector的信息会不会减少对kafka的依赖

可以有 把它做成可配置的形式 默认webhook

@hertzbeat
Copy link
Contributor

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


In the case of a small amount of monitoring, is the original mode retained to simplify the difficulty of deployment? At the beginning, it was only used when looking at hz without additional deployment of agent

Yes, the default is still all in one. The design will also minimize external component dependencies

If webhook is provided in the main hz to receive information from other collectors, will it reduce the dependence on kafka

It can be made into a configurable form, the default webhook

@l646505418
Copy link
Contributor

感觉这个更像是提供一个控制中心呢😂

@hertzbeat
Copy link
Contributor

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


It feels more like providing a control center😂

@tomsun28 tomsun28 changed the title hertzbeat metrics collector cluster design [Task] hertzbeat metrics collector cluster design Jul 13, 2023
@tomsun28 tomsun28 added enhancement New feature or request good first issue Good for newcomers task labels Jul 13, 2023
@tomsun28 tomsun28 linked a pull request Jul 13, 2023 that will close this issue
3 tasks
@tomsun28 tomsun28 moved this to Done in hertzbeat-v1.0 Sep 14, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
doc Improvements or additions to documentation enhancement New feature or request good first issue Good for newcomers new feature task
Projects
None yet
Development

Successfully merging a pull request may close this issue.

5 participants