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 problemsIf WordPress keeps sending you to /wp-admin/upgrade.php, or alternates between “Database Update Required” and “No update required, your WordPress database is already up to date,” stop refreshing and work through the problem in order. Run the database update once, clear every cache layer, temporarily bypass wp-content/object-cache.php, disable plugins, and then verify that the active WordPress files, database connection, and database permissions belong to the same site. Only consider database repair after a verified backup; never edit db_version just to hide the warning.
Antihost offers managed WordPress hosting with free migration, one-click staging and a 14-day free trial.
What the WordPress database-update loop means
WordPress stores the database schema level in the db_version option. The loaded WordPress core files contain the schema version they expect. During the normal update flow, /wp-admin/upgrade.php compares those values. If they differ, WordPress displays the upgrade screen and calls wp_upgrade() to run the required migrations.
As an Amazon Associate I earn from qualifying purchases.
Seeing the screen once after a core update is normal. A loop is different: one request decides that an upgrade is needed, another reports that no upgrade is needed, and the next request asks again. That usually means the migration did not persist, a request is reading a stale cached value, or different requests are reaching different code or database environments.
The wording on the core screen is intentionally cautious: “WordPress has been updated! Next and final step is to update your database to the newest version.” It also warns that “The database update process may take a little while, so please be patient.” Let the first run finish before diagnosing a problem.
Do these safety checks before changing files
A loop can be caused by a cache and may disappear without any destructive action. Prepare a way back before you change runtime files or database settings.
- Confirm you can access the hosting file manager, SFTP, or another file-management method.
- Make a copy of any file you plan to rename, especially
wp-content/object-cache.php. - Create and verify a database backup before any repair operation or direct database change.
- Record whether the site is production, staging, or part of a multi-server deployment.
- Keep a second browser window or an incognito window available for testing without your logged-in browser cache.
Do not repeatedly submit the upgrade form while a migration is still running. Repeated requests can make it harder to tell whether the original operation completed.
Fix the loop in the lowest-risk order
1. Run the normal database update once
- Open
/wp-admin/upgrade.phpfor the affected site. - Click the database-update button once.
- Wait for the completion message. Large sites can take time, and the WordPress screen explicitly asks administrators to be patient.
- When the process finishes, open a new incognito window and test
/wp-admin/.
If the dashboard opens normally, do not make further changes. If the same prompt returns, continue with cache diagnosis rather than manually changing the database version.
2. Remove stale browser, page, CDN, and plugin cache responses
A cached admin response can replay an old decision even after the database has changed. Test in a private window first; this separates a browser session problem from a server-side problem.
- Open the login and dashboard in an incognito or private window.
- Purge the site’s full-page cache, if one is enabled.
- Purge any CDN or reverse-proxy cache in front of the site.
- Clear cache data managed by WordPress caching plugins.
- Log in again and revisit
/wp-admin/upgrade.php.
Do not assume that clearing a page-cache plugin also clears an external object cache. Those are separate layers and can retain different values.
3. Temporarily bypass object-cache.php
WordPress can load a persistent object-cache drop-in from wp-content/object-cache.php. If that file returns an old db_version value, the update may succeed in the database while later requests continue to believe that the old schema is active.
- Use your hosting file manager or SFTP to open the site’s
wp-contentdirectory. - Make a backup copy of
object-cache.php. - Rename the live file to a temporary name such as
object-cache.php.disabled. Renaming is preferable to deleting because it is easy to reverse. - Open a private browser window and run the upgrade screen again.
- If the dashboard now works, purge the remaining cache layers and identify which cache service supplied the stale value.
Renaming the drop-in disables the object-cache integration while it is absent. A current hosting guide updated August 29, 2025, documents this diagnostic for the exact symptom and warns that Memcached functionality remains disabled until the cache is re-enabled. Do not leave the file disabled permanently unless you have deliberately chosen to run without that cache.
4. Disable plugins without deleting them
A plugin can cache admin responses, hook into the upgrade request, or depend on a schema state that has not finished changing. Disable plugins only long enough to test.
Rank #3
- Rename the
wp-content/pluginsdirectory to a temporary name such asplugins.disabledusing the file manager or SFTP. - Try
/wp-admin/upgrade.phpin a private window. - If the loop stops, restore the directory name to
plugins. - Re-enable plugins in small groups, testing the dashboard after each group.
- When the loop returns, narrow that group to the individual conflicting plugin and check its compatibility or cache settings.
This method leaves plugin files and settings intact. It also changes site behavior while plugins are unavailable, so perform the test during a maintenance window when possible.
5. Verify that the code and database belong together
A deployment mismatch can make the version check appear random. One web node may run newer core files while another serves older files, or a staging copy may be connected to a production database.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
- Check that every active server or deployment node contains the same WordPress core version.
- Open
wp-config.phpand verify the database name, host, user, and password point to the intended site. - Confirm the table prefix in
wp-config.phpmatches the tables that contain this site’s data. - Check that the account can write to the WordPress database, including the tables used by the installation.
- Confirm that the URL you are testing resolves to the same environment whose files and database you inspected.
Look at the PHP error log, web-server log, and hosting deployment log for failed SQL statements, permission errors, connection failures, or requests routed to another node. A migration cannot become durable if the database account cannot write it.
6. Use database repair only after a verified backup
Repair is a later diagnostic, not the first response to the message. If cache and plugin isolation do not help, create a fresh backup and confirm that it can be restored before proceeding.
- Back up the database and confirm the backup contains the WordPress tables.
- Use the repair procedure provided by your host or WordPress installation instructions.
- Keep the repair setting temporary and restricted to the repair window.
- Run the repair, review its result, and test the dashboard.
- Remove the temporary repair setting immediately afterward.
Do not hand-edit the db_version option to make the prompt disappear. That can mark an incomplete migration as complete and leave the schema inconsistent with the code that will use it.
7. Restore caching deliberately after access returns
Once /wp-admin/ works without the prompt, restore the original object-cache.php file or re-enable the cache plugin. Then purge every cache layer again and test in a private window. If the loop returns immediately after restoring the drop-in, the cache service is still serving an old value or is not compatible with the current deployment; leave it bypassed while you correct that configuration.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Match the symptom to the most likely cause
| What you observe | Most likely explanation | First action |
|---|---|---|
| The update says it succeeded, but the prompt returns on the next request. | A stale persistent object cache or another cache is returning an old db_version. |
Purge caches, then temporarily rename wp-content/object-cache.php. |
| Private browsing works, but the normal browser does not. | A browser session or cached response is stale. | Clear browser data and server-side caches; keep testing privately. |
| The loop stops when all plugins are disabled. | A plugin is interfering with the upgrade request or caching an admin response. | Restore the plugins directory and re-enable plugins in groups. |
| Different requests show different WordPress versions or site content. | Requests are reaching different deployment nodes, or the site is connected to the wrong database. | Compare core files, wp-config.php, table prefix, and routing. |
| Logs show SQL permission or connection errors. | The migration cannot persist because the database account or connection is failing. | Correct credentials and permissions, then rerun the normal update. |
| The loop remains after cache and plugin isolation. | A failed migration, damaged tables, or an environment mismatch needs deeper investigation. | Back up first, inspect logs, and use a controlled repair procedure. |
Understand the risk of each remedy
| Remedy | Reversible? | Runtime impact | Risk level |
|---|---|---|---|
| Private-window test and cache purge | Yes | Visitors may briefly receive uncached pages. | Low |
Temporarily rename object-cache.php |
Yes, if a backup was made | Persistent object caching, including Memcached integration, is disabled. | Low to medium |
| Temporarily rename the plugins directory | Yes | Plugin-provided features stop until the directory is restored. | Medium |
| Database repair | Not always | Repair may alter tables or indexes. | High without a verified backup |
Manual edit of db_version |
Technically possible, but unsafe as a shortcut | Can conceal unfinished migrations and create later failures. | Very high |
Why deleting object-cache.php is not the same as fixing the cache
The drop-in is a loader that lets WordPress use a persistent object-cache backend. Removing or renaming it can prove that cached data is involved, but it does not repair the backend, flush every stored key, or explain why the stale value was retained. Keep the original file, record its owner and permissions, and ask the host or cache administrator to flush or rebuild the relevant cache after the site is stable.
If you deleted the file without a backup, the cache plugin or hosting control panel may be able to regenerate it. Do not download a replacement from an unrelated site or copy one from another WordPress installation; drop-ins can contain site-specific connection settings.
Deployment and database checks that prevent a repeat
Check every web node
On a load-balanced site, compare the WordPress core files and deployment timestamp on every node. If one node still has an older release, a request can alternate between schema expectations. Make the deployment atomic where possible, or take the site out of rotation while the update runs.
Check the database target
Compare the database name, host, table prefix, and site URL in the active configuration with the environment you intended to update. A staging configuration pointed at a production database, or the reverse, can make an apparently successful update look as though it vanished.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallOutdated 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 matchCheck write permissions and failures
Successful page loads do not prove that migrations can write. Review database grants and the logs generated at the exact time of the upgrade. Correct a denied ALTER, CREATE, UPDATE, or connection operation before trying the upgrade again.
Validate the recovery before declaring it fixed
- Open
/wp-admin/in a private window and confirm the dashboard loads. - Visit
/wp-admin/upgrade.phpand verify that it no longer demands the same update. - Sign out, sign back in, and test from the normal browser after clearing its stale data.
- Purge page, CDN, plugin, and object caches in the order used by your hosting setup.
- Restore the original object-cache drop-in or re-enable the cache plugin, then repeat the private-window test.
- Re-enable plugins in groups if they were disabled, testing the dashboard after each group.
- Remove any temporary repair setting and confirm that your backup and change records are retained.
A successful recovery is not merely a screen that disappears once. It is consistent dashboard access across fresh sessions, with the same code, database, and cache configuration continuing to agree on the schema version.
Common mistakes that make the loop worse
- Editing
db_versionfirst: this hides the comparison instead of running a missing migration. - Deleting cache files without a copy: you lose the ability to restore the original integration and may disable performance features unexpectedly.
- Clearing only the browser cache: an external object cache or CDN can continue serving the stale value.
- Disabling plugins and forgetting to restore the directory: the dashboard may return while site features remain offline.
- Running repair without a tested backup: a repair cannot be safely reversed if it changes a damaged table.
- Testing only one request: a load balancer or inconsistent deployment can make the next request reach a different environment.
When to involve your host or developer
Escalate when you cannot identify the active database, when multiple web nodes disagree, when logs show repeated SQL failures, or when the loop continues with caches and plugins bypassed. Give the administrator the exact time of an attempted upgrade, the WordPress core version, the database host and name (without passwords), the table prefix, relevant error-log entries, and the steps already completed. This lets them trace routing, permissions, and cache invalidation without asking you to make risky direct edits.
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.
Recommended Free Tools
