how-to, ref: fix format and add a body heading (#2186)

how-to/configure/placement-rules.md

@@ -95,43 +95,43 @@ You can use pd-ctl to disable the Placement Rules feature and switch to the prev
 pd-ctl config placement-rules disable
 ```
 
-> **Notes:**
+> **Note:**
 >
 > After disabling Placement Rules, PD uses the original `max-replicas` and `location-labels` configurations. The modification of rules (when Placement Rules is enabled) will not update these two configurations in real time. In addition, all the rules that have been configured remain in PD and will be used the next time you enable Placement Rules.
 
 ### Set rules using pd-ctl
 
-> **Note：**
+> **Note:**
 >
 > The change of rules affects the PD scheduling in real time. Improper rule setting might result in fewer replicas and affect the high availability of the system.
 
 pd-ctl supports using the following methods to view rules in the system, and the output is a JSON-format rule or a rule list.
 
-- **To view the list of all rules:**
+- To view the list of all rules:
 
     {{< copyable "shell-regular" >}}
 
     ```bash
     pd-ctl config placement-rules show
     ```
 
-- **To view the list of all rules in a PD Group:**
+- To view the list of all rules in a PD Group:
 
     {{< copyable "shell-regular" >}}
 
     ```bash
     pd-ctl config placement-rules show --group=pd
     ```
 
-- **To view the rule of a specific ID in a Group:**
+- To view the rule of a specific ID in a Group:
 
     {{< copyable "shell-regular" >}}
 
     ```bash
     pd-ctl config placement-rules show --group=pd --id=default
     ```
 
-- **To view the rule list that matches a Region:**
+- To view the rule list that matches a Region:
 
     {{< copyable "shell-regular" >}}
 

reference/system-databases/sql-diagnosis.md

@@ -15,6 +15,8 @@ The SQL diagnosis system has the following advantages:
 + It provides monitoring summaries and automatic diagnosis.
 + You will find it easier to query cluster information.
 
+## Overview
+
 The SQL diagnosis system consists of three major parts:
 
 + **Cluster information table**: The SQL diagnosis system introduces cluster information tables that provide a unified way to get the discrete information of each instance and node. This system fully integrates the cluster topology, hardware information, software information, kernel parameters, monitoring, system information, slow queries, statements, and logs of the entire cluster into the table. So you can query these information using SQL statements.