0X000019E5

Fix ERROR_COULD_NOT_RESIZE_LOG (0x000019E5) on Windows

Windows Errors Intermediate 👁 0 views 📅 May 27, 2026

Stop wasting time with generic fixes. Here's the real cause and solution for this log resizing failure on Windows Server and 10/11.

Yeah, 0x000019E5 is annoying. It usually pops up when you're trying to manage event logs or a database hits its transaction log limit. The error means Windows can't grow a log file because the volume or the log's own metadata won't cooperate.

The Fix: Shrink and Compact the Log

Skip all the disk cleanup nonsense for now. What's actually happening here is the log file has internal fragmentation or a corrupt allocation bitmap. Run this in an elevated command prompt:

wevtutil sl Application /ms:20971520

That sets the Application log to 20 MB. If you're dealing with a different log—say Security or System—swap Application for the right name. Use wevtutil el to list all logs if you're not sure.

If the error's coming from SQL Server or some other app, the fix is similar: shrink the log file manually using that app's tools, then grow it again. For SQL Server, that's:

DBCC SHRINKFILE (YourLogFile, 100)  -- 100 MB target

Then set a new size with ALTER DATABASE ... MODIFY FILE. The reason step 3 works is simple: shrinking the log forces SQL to rewrite the internal pointers, clearing out any bit-level corruption that prevents resizing.

Why This Happens

Under the hood, every log file has a header that stores the current size and allocation info. When you request a resize, Windows checks if the volume has enough contiguous free space. But the real gotcha is the log's own internal allocation bitmap—if it's corrupt, the OS sees the space as “used” even when it's not. This is common after a sudden power loss or a failed disk write. The wevtutil sl command rebuilds that bitmap from scratch.

Less Common Variations

NTFS Transaction Log (NTFS $LogFile)

If you see 0x000019E5 during a chkdsk or a volume mount, it's the NTFS transaction log that's stuck. The fix is to run:

fsutil resource setlogsize C: 0

That resets the log to its default size. Warning: This disables the USN journal and can affect volume shadow copies. You'll need to re-enable them after:

fsutil usn createjournal m=1000 a=100 C:

Registry Log (Hive Transaction Log)

Rare but nasty. The Registry uses transaction logs for hive files like SAM or SYSTEM. If 0x000019E5 appears during a registry operation, the hive's log is full. Boot from a Windows Recovery Environment and run:

reg save HKLM\SYSTEM C:\system_bak.hiv

Then delete C:\Windows\System32\config\SYSTEM.LOG1 and SYSTEM.LOG2. This forces Windows to create fresh logs on next boot. Do this for SAM or SECURITY too if needed.

Prevention

Log resize errors love three things: low disk space, sudden power loss, and oversized logs that never rotate. Set a sane maximum log size—20 MB for most Windows event logs is plenty. For SQL Server, stick to a fixed size per log file, and grow in 100 MB increments, not by percent. Also, enable the NTFS transaction log's automatic growth setting via fsutil behavior set DisableLastAccess 1—that reduces metadata churn.

And if you're on a VM, check that your hypervisor isn't truncating writes (VMware's “Write Same” bug has caused this exact error on Server 2019). Update to the latest VM tools.

One last thing: don't run wevtutil cl blindly. That clears the log but doesn't fix the bitmap—you'll still get the error. Use wevtutil sl with a size parameter instead.

Was this solution helpful?