(Republishing, or using this info in a commercial product/website, is prohibited without permission. All other uses are permitted. If in doubt, please ask.)
This wait type occurs when a thread needs to issue an I/O, but all I/O for the database has been paused (using the undocumented DBCC FREEZE_IO command) by an external process that is backing up the database files directly, instead of using the VDI interface to perform a ‘SQL Server’ backup.
(Books Online description: “Occurs when a task is waiting to access a file when an external backup is active. This is reported for each waiting user process. A count larger than five per user process may indicate that the external backup is taking too much time to finish.”)
Questions/comments on this wait type? Click here to send Paul an email, especially if you have any information to add to this topic.
Added in SQL Server version:
Removed in SQL Server version:
Extended Events wait_type value:
The map_key value in sys.dm_xe_map_values is 148 in 2008 and 2008 R2, and 152 in 2012 and 2014 RTM. After 2014 RTM, you must check the DMV to get the latest value as some map_key values have changed in later builds.
There will be one wait per second, lasting for a second, for each thread that is waiting to perform I/O until I/Os are allowed again (using the undocumented DBCC THAW_IO command). The Books Online advice below is really unhelpful, as the time that I/O is frozen for will depend on how big the files are that are being backed up (e.g. for multiple terabytes of files, they’re obviously not going to be backed up in 5 seconds or less).
Known occurrences in SQL Server (list number matches call stack list):
- Waiting for I/O to be unfrozen (in this case, while performing readahead during an index scan)
And other similar call stacks.
Abbreviated call stacks (list number matches known occurrences list):