To retain this information for a longer period, copy the data to a separate database. The data in this view is retained for a maximum of 30 days, or possibly less depending on the number of databases and the number of unique events each database generates. During that time, the information within a single row may be updated periodically. Typically, the data is accumulated within an hour of the start of the aggregation interval, but it may take up to a maximum of 24 hours for all the data to appear in the view. For example, an event occurring exactly at 19:25:00.0000000 would be included only in the second interval shown below: start_time end_timeĭata in this view is accumulated over time. For these events, event_count will be 1 and start_time and end_time will equal the actual UTC date and time when the event occurred.įor example, if a user fails to connect to database Database1, because of an invalid login name, seven times between 11:00 and 11:05 on (UTC), this information is available in a single row in this view: database_nameĪn event is included in an aggregation interval when the event occurs on or after start_time and before end_time for that interval. Some events, such as deadlocks, are not aggregated. The event_count column represents the number of times a particular event_type and event_subtype occurred for a specific database within a given time interval. Remarks Event aggregationĮvent information for this view is collected and aggregated within 5-minute intervals. Users with permission to access the master database on the logical server in Azure SQL Database have read-only access to this view. User requested to change the password of the user logging in which is not supported.Ĭlient IP address is not allowed to access the server. User requested to attach a database file, which is not supported. Windows logins are not supported in this version of SQL Server. Login name is not valid in this version of SQL Server. Additional categories, event types, and subtypes may be added in future releases of SQL Database. This view does not include all possible SQL Database database events that can occur, only those listed here. The following table lists the types of events that are collected in this view.įor events in the connectivity category, summary information is available in the sys.database_connection_stats view. The events recorded by each row in this view are identified by a category ( event_category), event type ( event_type), and a subtype ( event_subtype). This column is unused and is preserved for backwards compatibility. The number of times that this event occurred for the specified database within the time interval specified ( start_time and end_time). See Event Types for a list of possible values. The high-level component that generated this event. For events that are not aggregated, start_time and end_time equal the actual UTC date and time of the event. For aggregated events, End_time is always exactly 5 minutes later than the corresponding start_time in the same row. UTC date and time of the end of the aggregation interval. For aggregated events, the time is always a multiple of 5 minutes. UTC date and time of the start of the aggregation interval. If the connection fails and the user did not specify a database name, then this column is blank. The sys.event_log view contains the following columns. To reduce the impact of this issue, limit queries of sys.event_log. For logical servers with a large number of databases and/or high numbers of logins, querying sys.event_log can cause high resource usage in the master database, possibly resulting in login failures.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |