soander commented on PR #9620: URL: https://github.com/apache/skywalking/pull/9620#issuecomment-1296405145
> I still noticed there are several mentioned things not been fixed. > > 1. There is no unit test to cover telegraf format converting. `TelegrafServiceHandler` is a HTTP handler. What it received is an HTTP text, what it looks like, and how it could be converted to SampleFamily should be well tested, including legal text or illegal text cases. > 2. You don't explain what is the relationship between this VM monitoring and OpenTelemetry's Linux monitoring. > 3. In [backend-vm-monitoring.md](https://github.com/apache/skywalking/pull/9620/files#diff-9b751b71ba658be11f59887b6462a97bd3675a181c3cd3fcf7d5c4e3d452c295), you mentioned, there are two ways to support VM monitoring. But they seem not the same, such as not using the same metric name. So, how the dashboard would work? In the document, there is not any document mentioning that. > > I still can't follow the whole picture of this feature implementation. Could any of you explain it? We can't deliver this to the community with this confusion. Especially, `telegraf-rules/vm.yaml` is included as a part of the official release. > > All default activated MAL scripts match with the dashboard settings, as well as the document description about the data flow. 1. I will write the HTTP text converting to SampleFamily unit test. 2. Telegraf and OpenTelemetry are two different metrics collection methods, both of them can collect linux system metrics and display it on the UI. 3. The metrics are collected by telegraf can also be displayed on the VM monitoring dashboard. The default telegraf MAL will convert the telegraf metrics into VM metrics. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
