Seeking guidance on how to configure Grafana monitoring items

Translated
1

Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: 求大佬指导下grafana 监控项如何配置

| username: cheng

[TiDB Usage Environment] Production Environment / Testing / POC
[TiDB Version]
[Encountered Issues]
[Reproduction Path] What operations were performed to encounter the issue
[Issue Phenomenon and Impact]
The initial Grafana monitoring of the cluster has default monitoring alert items, but each one shows “no data” (the cluster has been running for a long time and there is monitoring data). When I click in, there is an error. I searched online but didn’t find detailed explanations, just mentions of needing to modify Metrics.
I still don’t understand how to do it.
How should the SQL in metrics be written? Can I directly use the alert rules from the documentation? Also, if I want to configure an alert for memory usage of a machine in the process, which monitoring item should I go into to set the alert?
TiDB Cluster Alert Rules | PingCAP Docs

image

image
image1380×636 51.9 KB

image
image1380×372 36.4 KB

image
image1077×627 30.6 KB

2
| username: xiaohetao | Original post link

Figure 1:

  1. If it’s a newly started cluster and there’s no data, just wait and see;
  2. If not, check whether the export process of each node is normal;

Figure 2:

  1. There might be incorrect variable configurations in the monitoring parameters, check the parameter configurations (check if there are invalid variables or incorrect variable parameters);

Figure 3:
I’m not very sure about this alert configuration either, consult other experts.

3
| username: Billmay表妹 | Original post link

Refer to it.

4
| username: cheng | Original post link

The article mentions debugging, but the provided URL does not open.

5
| username: cheng | Original post link

The alert items in the configuration file are the same as those configured in the official documentation, but this alert item is actually not present in Grafana. Could you provide a screenshot of the configuration process?

6
| username: cheng | Original post link

I am not looking to collect Prometheus data. I want to configure monitoring alerts - specifically alerts on Grafana.

7
| username: jansu-dev | Original post link
  1. First of all, what you are seeing is the alerts that come with Grafana, while the alerts mentioned in the TiDB official documentation are from Prometheus.
  2. You can ignore Grafana’s alerts because many of the alert thresholds in it are inaccurate. It is recommended to use the alert rules provided by TiDB officially (Prometheus generates the alerts, and they are available upon deployment. Alertmanager sends the alerts to humans and needs to be configured manually). In other words, as long as you configure Alertmanager, you will receive the alerts.
  3. How to configure Alertmanager: 自定义监控组件的配置 | PingCAP 文档中心
8
| username: cheng | Original post link

I see that the cluster is already configured, so what should I do next?

9
| username: jansu-dev | Original post link

Start from here and configure as needed → alertmanager的使用_altermanger-CSDN博客
image

10
| username: system | Original post link

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.