MySQL error 1034 means MySQL found a damaged or inconsistent index for a table: Incorrect key file for table 'table_name'; try to repair it. It most commonly affects MyISAM tables. First identify the table engine, make a backup, run CHECK TABLE, and use REPAIR TABLE while the server is running or myisamchk only after stopping mysqld. Do not apply these MyISAM procedures to an InnoDB table without confirming a version-specific recovery plan.
Need a server to try this on? Antihost offers Linux VPS plans with NVMe storage, full root access and a 14-day free trial.
What error 1034 means
Error 1034 is MySQL’s ER_NOT_KEYFILE condition, with SQLSTATE HY000. The complete diagnostic is usually:
As an Amazon Associate I earn from qualifying purchases.
ERROR 1034 (HY000): Incorrect key file for table 'table_name'; try to repair itA key file is the on-disk index associated with a table. MySQL is reporting that the index does not agree with the table’s data, is damaged, or cannot be read consistently. The message is not a generic server failure: it identifies one table and tells you that repair may be required.
In practice, this error is strongly associated with MyISAM. MyISAM stores table data and indexes in separate files, so an interrupted write or an out-of-sync index can leave the table marked as crashed or not closed properly. The repair command, its risks, and the correct recovery path depend on the storage engine, so engine identification is the first step.
#1 Best Overall
Why a MyISAM key file becomes incorrect
Oracle MySQL documents several ways MyISAM corruption can occur:
- mysqld is killed while writing. A forced process termination can interrupt a data or index update.
- An unexpected shutdown. Power loss, a host reset, or an operating-system crash can leave files only partly updated.
- Hardware or storage failure. Bad memory, a failing disk, filesystem errors, or an unreliable storage path can write invalid data.
- External file modification. Copying, replacing, or editing table files while mysqld is using them can make the data and index inconsistent.
- A MySQL or MyISAM software bug. A repeatable failure after ordinary operation may indicate a defect rather than a one-off shutdown.
Repair fixes the affected table’s recorded structure; it does not repair the underlying cause. If the same table becomes corrupt again, continuing to run repair commands without investigating restarts, storage, access patterns, and logs risks repeated data loss.
Recommended Free Tools
Before changing anything: protect the table
Repair can discard damaged rows or rebuild an index from whatever data remains readable. MySQL explicitly warns that repair may cause data loss. Preserve a recoverable copy before running a command that changes files.
Make a logical backup when the server can still read the table
If the server is running and the table can be selected, export the table or the relevant database using your normal backup process. Keep the original backup separate from any repaired copy. A logical dump gives you a second way to recover rows if a repair removes records or changes a damaged row.
Preserve files for an offline investigation
For a filesystem-level investigation, stop writes and preserve the table’s files before attempting repair. Do not copy individual MyISAM files while mysqld is actively changing them. A filesystem snapshot or a full datadir backup taken with the server stopped provides a point-in-time copy for a second attempt or specialist recovery.
Record the incident
Note the table name, the exact error, the time it appeared, recent restarts, and any storage or operating-system alerts. Save the MySQL error log. These details are important if corruption returns or you need to compare the original and repaired table.
Free tools Windows power users keep installed
One-click scans. No signup required.
Step 1: Identify the storage engine
Run one of these statements, replacing the example name with the affected table:
SHOW TABLE STATUS LIKE 'table_name';Look for the Engine column. You can also inspect the complete definition:
SHOW CREATE TABLE table_name;If the engine is MyISAM, the checks and repairs in the next sections apply. If it is InnoDB, stop and use the InnoDB-specific guidance below; REPAIR TABLE and myisamchk are MyISAM procedures.
Step 2: Check a MyISAM table
For a MyISAM table, ask MySQL to inspect the table and its indexes:
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Clear out junk files and repair common Windows errorsFree Scan →CHECK TABLE table_name;The result normally includes a status such as OK or Table is already up to date when no repair is needed. If the result reports errors, corruption, or an improperly closed table, MySQL’s guidance is normally to repair it. Save the complete result before proceeding.
A table may still be marked as “corrupted” or “not closed properly” even when the check finds no current structural problem. MySQL notes that CHECK TABLE can remove that mark when the check finds no problems. In that case, retry the operation that produced the warning and monitor the error log rather than immediately forcing an offline repair.
Repair a MyISAM table with SQL
When mysqld is running and the table is accessible, the SQL route is the safest first repair attempt because MySQL performs the operation through the server instead of an external process changing a live table.
REPAIR TABLE table_name;Read every row returned by the statement. The result reports the table, operation, message type, and message text. A successful repair should be followed by another check:
CHECK TABLE table_name;Then test the application query that originally failed and verify important row counts or application records against your backup. A successful command does not prove that every damaged row was recoverable; it proves that MySQL completed the requested repair operation.
When the SQL repair cannot run
The statement may fail if the table cannot be opened, the server lacks the required access, the table is not MyISAM, or the damage is too severe for the normal repair algorithm. Do not switch directly to myisamchk while mysqld is still running. The command-line utility reads and writes the same table files, and concurrent server access can make corruption worse.
Offline repair with myisamchk
Use this path only for a MyISAM table and only after stopping the MySQL server cleanly. Oracle MySQL’s guidance is explicit: if you repair a table from the command line, stop mysqld first.
- Stop the MySQL service using the service manager appropriate to your operating system.
- Confirm that mysqld has exited and that no other MySQL instance is using the datadir.
- Change to the directory containing the affected table’s
.MYIindex file, or provide the correct path to that file. - Run the quick recovery command:
myisamchk -r -q table_nameThe command may need the table’s full file path, for example the path to table_name.MYI. Follow your installation’s ownership and permissions rules; do not run a second mysqld against the same files.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchPC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Escalate only when the previous level fails
If quick recovery cannot repair the index, run the ordinary recovery mode:
myisamchk -r table_nameIf ordinary recovery also fails, use the slower safe-recovery algorithm:
myisamchk --safe-recover table_nameSafe recovery can take substantially longer because it uses a more conservative method. Keep the original backup untouched while trying each level. After the command completes, run a check with the server still stopped if appropriate, restart mysqld, and execute CHECK TABLE table_name; through SQL before returning the table to application traffic.
Choosing myisamchk checks
The check mode determines how thoroughly the utility scans the table:
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
| Command | Purpose | Published detection figure or limitation |
|---|---|---|
myisamchk table_name |
Basic consistency check. | The MySQL 8.4 manual says it finds 99.99% of errors; this is a documentation figure, not a guarantee for a particular table. |
myisamchk -m table_name |
More extensive check. | The MySQL 8.4 manual says it finds 99.999% of errors; the figure is a documented detection estimate, not a repair-success rate. |
myisamchk -e table_name |
Complete extended check. | Can take a long time, especially for large tables; use it when a deeper offline examination is justified. |
These percentages describe the checks’ documented ability to find errors. They do not mean that the same percentage of corrupted rows will be recovered, and MySQL publishes no general repair-success or corruption-prevalence statistic for this error.
When a difficult MyISAM repair needs USE_FRM
Some damaged tables require a more involved procedure. MySQL documents a difficult-repair path that creates new empty data and index files, copies the old data file back, and then returns to the normal repair stage. In supported cases, the SQL form can perform that procedure automatically:
REPAIR TABLE table_name USE_FRM;Use this only after making a backup and confirming that the table is MyISAM and that ordinary recovery is inadequate. The option relies on the table definition file and is not a universal solution for every form of damage. Validate the result with CHECK TABLE, application queries, and a comparison with your backup.
Replication warning
Filesystem-level operations are not logged by MySQL. Stop replication before performing offline file replacement or other filesystem repair, then coordinate how the repaired table will be brought back into synchronization. Otherwise, a replica can continue using the old damaged files or silently diverge from the repaired source.
Do not use MyISAM repair commands for InnoDB
An InnoDB table has a different storage and recovery model. Do not assume that REPAIR TABLE or myisamchk can fix an InnoDB table that displays error 1034.
MySQL 5.1 release notes describe a historical InnoDB case in which error 1034 resulted from the .frm file and the InnoDB data directory becoming out of sync after a column rename followed by index creation. The documented workaround was to create another table with the same structure and copy the original table into it. That advice belongs to the specific historical failure and version context; it is not a universal modern InnoDB recipe.
For a current InnoDB incident, preserve backups, capture the exact server version and error-log messages, and follow a recovery plan appropriate to that version and the actual InnoDB damage. Avoid deleting or replacing InnoDB files based on a MyISAM procedure.
Verify the repair before reopening normal traffic
Use a staged verification sequence rather than relying on a single “OK” message:
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEFix the driver behind crashes, sound loss and screen glitchesFind Drivers →- Run
CHECK TABLE table_name;and save the output. - Run representative reads, including the query that originally failed.
- Compare key counts, recent records, and application-level totals with the backup or another trusted copy.
- Check the MySQL error log for new warnings after restart.
- Watch the table during normal writes for another occurrence of error 1034 or a “crashed” status.
If rows are missing, indexes remain inconsistent, or the error reappears immediately, stop writes and return to the preserved backup rather than repeatedly repairing the same files.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Stop recurring corruption
A one-time repair after an unexpected shutdown may be enough. Repeated corruption requires investigation.
Review shutdown and restart history
Correlate the first error with power events, host resets, out-of-memory kills, container termination, scheduled restarts, and forced service stops. If mysqld is repeatedly terminated during writes, repair is only treating the symptom.
Inspect storage and hardware
Check operating-system logs, filesystem health, disk error counters, controller alerts, and available space. A failing or intermittently disconnected storage device can corrupt multiple tables. Correct the storage fault before trusting another repair.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problemsCheck concurrent access
Find any backup, synchronization, migration, or administrative script that opens or modifies MyISAM files outside mysqld. External modification while the server is using a table is a documented corruption cause.
Read the MySQL error log
Look for the first warning, not only the later 1034 message. Earlier messages may identify an aborted write, an I/O error, or a failed shutdown. Keep the log with the table backup and repair output.
Consider automatic MyISAM recovery options
MySQL can automatically check and attempt repairs when myisam_recover_options is enabled, and it records checking and repair warnings. Automatic recovery can reduce downtime after a crash, but it does not replace backups or root-cause analysis. Review the resulting warnings and confirm that the option is appropriate for your version and operational policy before enabling or changing it.
A practical decision path
| Situation | Recommended next action | Do not do |
|---|---|---|
| Engine is MyISAM and the table is readable | Back up, run CHECK TABLE, then use REPAIR TABLE if errors are reported. |
Do not edit or replace table files while mysqld is serving them. |
| SQL repair fails or the table cannot open | Stop mysqld, preserve the files, and try myisamchk -r -q, then escalate only as needed. |
Do not run myisamchk against a live table. |
| Ordinary recovery fails | Use myisamchk -r, then --safe-recover if necessary; consider USE_FRM for a supported difficult-repair case. |
Do not overwrite the only copy of the original files. |
| Engine is InnoDB | Use an InnoDB- and version-specific recovery plan and preserve the error log and backups. | Do not assume MyISAM repair commands apply. |
| Corruption returns | Investigate shutdowns, storage, external access, logs, and software defects. | Do not treat repeated repair as a permanent fix. |
When to escalate
Escalate to an experienced database-recovery specialist or MySQL support when the table contains irreplaceable data, no verified backup exists, repairs remove rows, multiple tables are affected, or corruption returns after the underlying environment appears healthy. For a reproducible failure, MySQL’s documented procedure is to stop mysqld, make backups, check with myisamchk, repair affected tables, reproduce the problem with logging enabled, and provide the collected evidence to the MySQL Customer Support Center if you have a support agreement.
There is no authoritative general statistic showing how often error 1034 occurs or how often repair succeeds. The outcome depends on the engine, the extent and cause of damage, the quality of available backups, and whether the storage problem has been fixed.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
