Analyzing database problems with the Database tier
The Database tier in the Application View provides a visual indication about the database performance and availability compared to the set Minor and Critical thresholds. This topic describes how to use the Database tier to identify performance and availability problems in your application databases.
Before you begin
To perform this procedure, you must have Application Operator-level access, or higher.
Determine if the application has problems in the Database tier
In the Application View, the Database tier displays the number of databases that have one or more problems (latency violations or errors) with queries or database operations during the selected period. The tier displays the number of database instances that impact one or more transaction, causing latency violations or errors:
- Impacted Databases (number inside the circle). Number of impacted database instances
- Colored part of the circle. Ratio of impacted database instances to total database instances. The color reflects whether the percentage of databases experiencing issues exceeds the defined thresholds.
- Operations per second. Number of database queries and operations per second
The following example describes the information in the Database tier:
For the five-minute period reflected in the example, the Database tier shows that the combined number of operations per second for all the databases in the tier is three operations per second. The number of impacted databases, that is, the number of databases with latency violations or errors on queries or database operations, is two. You can see that the number of impacted databases is approximately a quarter of all the servers in the tier.
To analyze database problems with the Database tier
- On the Application View tab of the Application Details page, position the time slider to select a five-minute period of interest.
- Click the Database tier to display the databases that were active in the time period.
Example of the Tier Members and Selection Details
In the Tier Members section of the page, click a filter to show or hide databases with the selected severity.
Filters for databases with Critical and Minor events are selected by default to show the servers with detected issues.
Evaluating details of selected databases
Click a database name to select a database (or to clear the selection) and examine details about problems with the database.
- Examine a summary of database issues
- Examine specific problems with databases
- Examine the slowest database queries
Examine a summary of database issues
For selected servers, the following details are displayed on the Summary tab, below the Tier Members section. Click a column heading to sort entries by that category.
Column | Description |
|---|---|
Action menu | Click to select View Business Transactions to see the transactions that queried the selected database |
Severity icon | Representation of the severity level of events on the database |
Database Name | Depending on the type of database, the name is displayed using the following syntax:
|
Operations/Sec | Number of database queries and operations sent to the database per second |
Impacted Transactions | Percentage of database operations with latency violations or errors that exceed defined thresholds |
Latency Violations | Percentage of database operations with latency violations |
Errors | Percentage of database operations server with errors |
To examine transaction details for a selected database, from the Summary tab, select View Business Transactions from the row action menu of a specific database.
Examine specific problems with databases
Click the Problems tab to examine the types of problems on the selected databases.
The following details are displayed about a database problem. Click a column heading to sort entries by that category.
Column | Description |
|---|---|
Severity icon | Representation of the event severity level |
Problem | Brief problem description, such as Operation latency on the database or Operations with errors on the database. |
Database Name | Name of the database The severity level remains unchanged until all problems with operations of that level are closed. |
Problem Started | Time and date that the problem started which caused the displayed severity If the problem started earlier than the displayed time, click the start time to reset the time selection. The time slider moves to the five-minute period that includes the problem start time. |
Threshold | Value of the threshold that was crossed For a list of default threshold values, see Application-server-metrics. |
Current Status | Open or Closed, which represents the current status of the problem, and time in which the problem was resolved, if applicable |
Examine the slowest database queries
Analyze the slowest database queries on the selected databases. A database can show no problems, or no latency violations, yet you can still examine the slowest queries.
The Top 10 Slowest Queries lists the ten slowest queries for the selected database member or members, for the selected five-minute period. No matter how many databases are selected, only ten queries are listed, sorted according the the maximum latency by default. The slow-queries list provides the following information for each query:
- Query name
- Number of slow requests, that is, the number of requests that exceeded a performance threshold
- Total number of requests
- Average latency, in milliseconds
- Maximum latency, in milliseconds
- Minimum latency, in milliseconds
- Database name
Where to go from here
Analyzing-business-transactions
Related topics
Setting-up-applications-for-monitoring
Analyzing-end-user-experience-with-the-User-tier
Investigating-problems-with-Web-and-Business-tiers
Troubleshooting: Same-database-instances-appears-twice-in-Application-View