Assume that you use a Microsoft SQL Server 2014 or 2016 Always On Availability Group (AG). If a write error that resembles the following occurs on a secondary database, the database might become suspended.
Error: 17053, Severity: 16, State: 1.
SQLServerLogMgr::LogWriter: Operating system error 6(The handle is invalid.) encountered.
Write error during log flush.
In this situation, if you resume the data movement, the database isn't resumed, and it remains in the suspended state.
To work around this issue, you may restart the SQL Server instance, or you may remove the secondary database from the availability group and then add it back.
This issue is fixed in the following cumulative update for SQL Server:
After you install this update, if this issue occurs again, you may run the following Transact-SQL command to restart the database and then resume the data movement for the database.
ALTER DATABASE <database_name> SET HADR RESUME
This isn't automated. Therefore, you have to manually issue the resume operation. Because most types of errors that result in suspension of data movements on the secondary replica require manual intervention.
For example, if the log file is located on a shared folder or stored in Microsoft Azure Blob, and the connection is lost, the Error 17053 will occur. The manual intervention will make sure that the connection to the shared folder or the Azure Blob is restored first before you issue the HADR resume command.
Microsoft has confirmed that this is a problem in the Microsoft products that are listed in the "Applies to" section.