Audit Logging
Last updated
Was this helpful?
Last updated
Was this helpful?
Audit logging in Servoy enables tracking and recording of key events related to database tables, including data changes (inserts, updates, deletes) and data access (selects). This information is stored in a designated log table, which can reside in the same or a separate database server.
By configuring audit logging, developers can:
Ensure Compliance: Meet regulatory requirements like the HIPAA Security Rule by tracking data views and changes.
Monitor Data Changes: Capture and analyze changes to the data for debugging, operational insights, or analytics.
Use Audit Log Data: Build forms against the log table, allowing users to review historical data changes or access modified records.
The log table is a standard database table, and its structure provides detailed tracking, including timestamps, action types (e.g., insert, update, delete, view), affected tables and columns, and user identifiers. Developers can also configure client statistics logging to track user connections and disconnections, enhancing system monitoring.
Performance Note: Logging introduces a performance impact, especially for high-traffic solutions. Using a separate database for the log table can help mitigate this impact by optimizing for insert operations and enabling independent backup or archiving strategies.
To enable and configure audit logging in Servoy, follow these steps:
Set up a Log Server:
Go to the database server configuration page in Servoy Developer.
Select a database server to act as the log server, check the Log Server option and save the change
Click the Create Log Table button to generate a log table in the selected database server.
Enable Table Tracking:
Open the desired table in the table editor.
Navigate to the Security tab.
Under the selected group's table settings, enable:
Tracking (Insert/Update/Delete) to log changes such as adds, edits, or deletes.
Tracking (Select) to log data reads or views.
Log Configuration Recommendations:
Use a separate database for the log table if performance is critical.
Optimize the log database for insert operations, as most audit logging involves adding new records.
Plan a strategy for backing up or archiving large log tables.
The audit log data is stored in a table called log, which is a standard database table. Developers can build forms or scripts to interact with this data, making it accessible for reporting or troubleshooting.
Below are the columns in the log table and their descriptions:
event_time
The timestamp of the event (insert, update, delete, or select).
log_id
A unique, auto-incrementing identifier for the log record.
log_action
The action type: 1=delete, 2=insert, 3=update, 4=view.
server_name
The name of the database server where the action occurred.
table_name
The name of the affected table.
column_name
The name of the affected column (if applicable).
pk_data
The primary key data of the affected record.
old_data
The value of the data before the change (null for views).
new_data
The value of the data after the change or the viewed data.
user_uid
The unique identifier of the user who performed the action.
Servoy also logs client connections and disconnections in a table called client_stats.
This table can be created by clicking the Create Client Statistics Table button in the log server editor.
By default, this feature is enabled and tracks the following information:
pk_id
A unique identifier for the log entry.
server_ip
The IP address of the Servoy Application Server.
server_name
The hostname of the server.
total_clients_running
The number of active clients at the time of the log entry.
client_id
A unique identifier for the client.
start_time
The timestamp when the client connected.
stop_time
The timestamp when the client disconnected.
extra_info
A column used to detect alterations to the logging data.
user_uid
The unique identifier of the logged-in user.
This logging is enabled by default but can be disabled in the admin page under "Admin Settings":
By default, servoy-sequence is used for key generation, but it can be customized to use UUIDs or database-sequence.
