What’s the issue?
SQL Server database files have an autogrowth setting that controls how the file expands when it runs out of space. Autogrowth can be configured with a fixed or percentage-based growth value, or it can be disabled entirely by setting the growth value to 0. When autogrowth is disabled, the file cannot grow beyond its current size, and any operation that would require additional space fails with an error.This finding identifies tempdb data or log files configured with autogrowth disabled. The setting is sometimes applied as a perceived safety measure to prevent tempdb from filling a drive, but it produces the opposite of the intended effect.
Why is this a problem?
Tempdb is shared by every database and every session on the instance and is involved in sorts, hash joins, spills, temporary tables, table variables, and version store activity. When tempdb runs out of space, queries fail with error 1105, sessions are disrupted, and in severe cases the entire instance can become unstable until space is freed.Disabling autogrowth on tempdb files creates an artificial ceiling that triggers these failures even when free space is still available on the drive. A single large query, an unexpected sort spill, or a long-running transaction holding the version store open can hit the cap and cause widespread query failures across all databases on the instance.
The intended safety benefit (preventing tempdb from filling the drive) is better achieved through proper drive sizing and monitoring than through a hard size limit. A capped tempdb shifts the failure from a drive-level event to a query-level event, but the failure still occurs and is often more disruptive because it surfaces as application errors during normal operation.
The condition is closely related to setting a maximum size on tempdb files, which produces similar problems and is covered separately. Both settings represent attempts to limit tempdb growth that produce more problems than they solve and are not recommended as defensive measures.
What should you do about this?
Enable autogrowth on each affected file with a fixed growth value of at least 64 MB (256 MB or 512 MB may be more appropriate choices). Apply the same growth value across all data files so they grow consistently and the proportional fill algorithm continues to work correctly.Pair this with proper initial sizing of tempdb files, ideally filling most of the dedicated tempdb drive at startup, so autogrowth events are rare in normal operation. Confirm Instant File Initialization is enabled to minimize the impact of any growth events that do occur, and place tempdb on dedicated, fast storage sized for the actual workload.