You're trying to open Event Viewer, or maybe a backup program's failing, and you get hit with 0x000019E3 — ERROR_LOG_TAIL_INVALID. The message says something about an archive tail or base being invalid. In plain English: your Windows event log files (.evtx files) are corrupt.
I've seen this happen most often after a hard crash — power outage, blue screen, someone yanking the plug. Had a client last month whose server shut down during a storm. Next morning, Event Viewer was dead, and their monitoring software kept spitting out this error. The fix took ten minutes.
Let's get into it. The fix depends on exactly what's broken, so I'll walk through the three most common causes, starting with the one that works 80% of the time.
Cause #1: Corrupt Event Log Registry Entries
The most common trigger: Windows keeps metadata about each log file in the registry under HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\EventLog. When that metadata gets out of sync with the actual .evtx file — say, after an improper shutdown — you get this error.
Here's the fix: Delete the corrupt log files, then clear the registry entries that point to them. But you need to know which log is the problem.
- Open Event Viewer (
eventvwr.msc). If it won't open at all, skip to step 3. - Look under Windows Logs. If you see a yellow warning or red error next to a log (like Application, System, Security), that's your culprit. Note the log name — case-sensitive.
- If Event Viewer won't open, check the System event log in a different way: open Command Prompt as admin and run
wevtutil gl Application(replaceApplicationwith the log you suspect). If it returns an error like "The log is corrupt", you've found it.
Once you know the log name:
- Open Registry Editor (
regedit) as admin. - Go to
HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Services\EventLog\<LogName>. For example,SystemorApplication. - In the right pane, look for these values:
File(points to the .evtx file) andMaxSize,Retention,AutoBackupLogFiles. - Back up this key — right-click the log name key, Export, save it somewhere safe. This is your 'oops' button.
- Delete the
Filevalue. Or, if you're bold, delete the entire log key (but then you lose settings like max size — I usually just delete the File value). - Close Regedit. Open an admin command prompt.
- Run
wevtutil cl <LogName>— this clears the log. If it fails, it's because the .evtx file is still referenced. Restart the Windows Event Log service: in Services.msc, find it, right-click, Restart. - Now try Event Viewer again. It should recreate the .evtx file fresh.
Real-world note: I've had this fail when the .evtx file itself is locked by a process. In that case, reboot into Safe Mode and repeat. The service won't start in Safe Mode, so you can delete the .evtx files directly from C:\Windows\System32\winevt\Logs.
Cause #2: Log File Size Limit Hit During Corrupt Write
Windows event logs have a max size — default is 20 MB for System and Application logs. If a write operation (logging an event) happens right when the file hits that limit, and the system crashes mid-write, the log can end up with a corrupt tail. The log thinks it's full, but the tail pointer doesn't match.
This is actually the same error as Cause #1, but the fix is simpler: just clear the log. No registry deletion needed.
- Open Command Prompt as admin.
- Run
wevtutil cl System(replace with the log name). - If that fails with "The log is corrupt", try
wevtutil al System— this archives the log (backs it up) and clears it. The backed-up file goes to%SystemRoot%\System32\winevt\Logs\Archive-<LogName>-<timestamp>.evtx.
If it still fails, you'll need to boot into Safe Mode or use a Windows PE environment to delete the .evtx files directly. But try the archive command first — I've seen it fix things that a straight clear wouldn't.
Pro tip: After fixing, increase the max log size to 50 MB or even 100 MB if you're on a production server. This reduces the chance of hitting the limit during a crash. Use wevtutil sl System /ms:102400000 (that's 100 MB in bytes).
Cause #3: Drive-Level Corruption or Bad Sectors
Less common, but when it happens, it's nasty. If your hard drive is developing bad sectors and the event log file sits on one, Windows may write partial data or read back garbage. The log tail gets corrupted.
How to tell? If you've tried the fixes above and the error comes back within days — or if other files on the same drive show similar corruption — suspect the drive.
Check the drive:
- Open an admin Command Prompt.
- Run
chkdsk C: /f /r(replace C: with the drive where Windows is installed). This will schedule a check on next reboot. - Reboot and let chkdsk run. It may take an hour or more on large drives.
- After it finishes, check Event Viewer again. If chkdsk found and fixed bad sectors, the log may be repairable.
If that doesn't work, you've got bigger problems. The event log files are probably toast. You'll need to:
- Boot from a Windows installation USB.
- Go to Repair > Troubleshoot > Command Prompt.
- Navigate to
C:\Windows\System32\winevt\Logs(drive letter may be D: in WinPE). - Delete all .evtx files. Yes, all of them. Windows will recreate them on next boot.
- Reboot normally. Event Viewer will show blank logs, but they'll start populating immediately.
Had a client with an aging Dell server whose RAID array had one failing disk. Event log corruption was the first symptom. Replaced the disk, rebuilt the array, and no more 0x000019E3. Moral: if this error keeps coming back, check your storage health.
Quick-Reference Summary Table
| Cause | Symptom | Fix | Time |
|---|---|---|---|
| Corrupt registry references | Event Viewer fails to open specific logs, error on startup | Delete File value in registry under EventLog key, then clear log with wevtutil | 10 min |
| Log size limit hit during crash | Error appears after unexpected shutdown | Archive or clear log with wevtutil; increase max size | 5 min |
| Bad sectors on drive | Error recurs after clearing logs; other files corrupt | Run chkdsk; if fails, delete all .evtx files from WinPE | 1-2 hours |
That's it. Most people only need Cause #1. If you're here, you're likely dealing with a system that's crashed hard. The event logs are the canary in the coal mine — fix them, but also check why the system crashed in the first place. A UPS and a stable disk subsystem would prevent 90% of these occurrences.