hertzbeat
hertzbeat copied to clipboard
apache
希望自定义监控的自定义权限更高些
1.很多开发人员，非专业的服务器运维人员，更多的时候是Free -h 后回去自己写正则等拼凑服务器，去用AWK 这些难度很大。希望可以将parseType 这里加个自定义，再可以用反射或者JS引擎 ，动态的解析命令，返回结果 2.采集的时候configmap 能不能写成一个配置数组或者map，协议那边直接写下标或者key去取，少点配置。 3.碰到一些综合的采集，比如先执行SSH,再执行SQL，甚至还会调一些HTTP请求的，这边能用反射拓展开放吗？
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
- Many developers, non-professional server operation and maintenance personnel, more often go back after Free -h and then go back to write their own regularities and other pieces of the server. It is very difficult to use AWK. I hope that you can add a custom parseType here, and then you can use reflection or JS engine to dynamically parse commands and return results
- Can the configmap be written as a configuration array or map when collecting, and the subscript or key can be directly written on the protocol side to get it, with less configuration.
- When encountering some comprehensive collections, such as executing SSH first, then executing SQL, and even calling some HTTP requests, can we use reflection to expand and open?
谢谢反馈很有用的建议
1.很多开发人员，非专业的服务器运维人员，更多的时候是Free -h 后回去自己写正则等拼凑服务器，去用AWK 这些难度很大。 希望可以将parseType 这里加个自定义，再可以用反射或者JS引擎 ，动态的解析命令，返回结果
这个可以有的，就是需要考虑怎么去调用实现解析脚本
2.采集的时候configmap 能不能写成一个配置数组或者map，协议那边直接写下标或者key去取，少点配置。
这个已经在计划的支持中，目前configmap是key-value，但不能满足一些加header等场景，争取尽快支持。
3.碰到一些综合的采集，比如先执行SSH,再执行SQL，甚至还会调一些HTTP请求的，这边能用反射拓展开放吗？
你说的是一个监控类型比如MYSQL，可能需要采集ssh监控主机CPU内存，还需要执行SQL查询库指标这种场景吗，目前是已经支持了，在监控定义yml里面对不同的指标配置不同的采集协议就可以的。
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
Thank you for your feedback. Useful suggestions
- Many developers, non-professional server operation and maintenance personnel, more often go back after Free -h and then go back and write regularities and put together the server. It is very difficult to use AWK. I hope that you can add a custom parseType here, and then you can use reflection or JS engine to dynamically parse commands and return results
This is possible, but you need to consider how to call the implementation parsing script
- Can the configmap be written as a configuration array or map when collecting, and the subscript or key is directly written on the protocol side to get it, and there is less configuration.
This is already planned for support. At present, configmap is key-value, but it cannot meet some scenarios such as adding headers, and strive to support it as soon as possible.
- When encountering some comprehensive collections, such as executing SSH first, then executing SQL, and even calling some HTTP requests, can we use reflection to expand and open?
You are talking about a monitoring type such as MYSQL, you may need to collect ssh to monitor the host CPU memory, and also need to execute SQL query database indicators. It is currently supported. In the monitoring definition yml, configure different collections for different indicators Agreement is fine.
你好，关于第三点意思是这样子的。 是一些新型数据库的指标，不是单纯的JDBC就可以采集的。需要ssh+jdbc一起执行才可以，先去SSH执行一些指令拿到结果，再去jdbc里面拿结果，这样子的。
Bot detected the issue body's language is not English, translate it automatically. 👯👭🏻🧑🤝🧑👫🧑🏿🤝🧑🏻👩🏾🤝👨🏿👬🏿
Hello, this is the meaning of the third point. It is an indicator of some new databases, not simply JDBC can be collected. You need to execute ssh+jdbc together, first go to SSH to execute some instructions to get the result, and then go to jdbc to get the result, like this.