Fix "Briefly Unavailable for Scheduled Maintenance"

The message “Briefly unavailable for scheduled maintenance. Check back in a minute.” is almost never a real outage. It comes from a single hidden file named .maintenance sitting in your WordPress root folder, and deleting that file brings the site back in seconds. The more useful question is why WordPress left it behind, because that answer points to a failed update, a PHP timeout, or a plugin that died halfway through its own install routine.

WordPress Writes One File, Then Forgets to Delete It

Every time you run an update from Dashboard > Updates, WordPress drops a .maintenance file next to wp-config.php before it touches any code. While that file exists, the wp_maintenance() function intercepts requests and returns a 503 Service Unavailable response with the short notice visitors see.

On a healthy server the file is created and removed within two to ten seconds per plugin, so nobody notices. The error only becomes visible when the update process stops before cleanup runs. At that point the site is fine, the database is fine, and one orphaned file is holding the door shut.

Why the Maintenance Lock Gets Stuck

In our support queue, roughly four out of five stuck maintenance screens trace back to an interrupted update rather than a broken site. The usual causes:

The browser tab was closed or the connection dropped mid-update, so the cleanup step never fired.
PHP hit max_execution_time, often the default 30 seconds, while unzipping a large plugin or theme.
Memory ran out during a bulk update of 15 or 20 plugins queued at once.
A fatal PHP error in the new plugin version killed the request before WordPress finished.
File permissions stopped PHP from deleting its own file, common after a sloppy migration that left root files owned by the wrong user.

A stuck screen that appears without any update activity usually means a maintenance-mode plugin was switched on, or a cached 503 response is still being served from the edge.

Delete the .maintenance File

This is the fix for the overwhelming majority of cases, and it takes under a minute. Work through your host’s file manager, SFTP, or SSH.

Using SFTP or a File Manager

Connect to your site and open the WordPress root, the folder holding wp-config.php, wp-content, and wp-admin.
Turn on show hidden files (in FileZilla: Server > Force showing hidden files), since the filename starts with a dot.
Delete .maintenance.
Hard refresh the front end with Ctrl+Shift+R or Cmd+Shift+R.

Using SSH or WP-CLI

If you have terminal access, two commands handle it:

cd ~/public_html && ls -la | grep maintenance to confirm the file exists.
rm .maintenance to remove it, or wp maintenance-mode deactivate if WP-CLI is installed.

SSH is also the fastest route when the dashboard itself is locked out, the same approach we use for diagnosing the WordPress white screen of death.

When the Notice Comes Back After You Delete It

A maintenance screen that reappears within minutes means something is still actively writing the file or serving an old response. Check these in order:

Page and object cache. A 503 captured by your cache layer or CDN will keep showing after the file is gone, so purge everything. Our LiteSpeed Cache setup clears both object and edge caches from one button.
Auto-updates retrying. If a plugin update fails repeatedly, WordPress core auto-updates will try again on the next cron run and re-create the lock.
A maintenance plugin. Rename its folder inside wp-content/plugins to disable it, then log in and check its settings.
Leftover upgrade folders. Clear out wp-content/upgrade and any plugin-name.tmp directories so the next update starts clean.
Host-side work. Genuine platform maintenance is announced, so read your provider’s live status page before digging further.

Reading the PHP Error Behind the Lock

If the same plugin keeps failing, the maintenance notice is a symptom and the PHP log holds the cause. Add define('WP_DEBUG', true); and define('WP_DEBUG_LOG', true); to wp-config.php, then reproduce the update and read wp-content/debug.log.

Timeout entries and “allowed memory size exhausted” lines both point at resource limits rather than buggy code. We run WordPress plans at 256MB memory and a 120-second execution limit for that reason, which is enough headroom for large page builder and WooCommerce updates. The same limits quietly fix unrelated issues like HTTP errors during image uploads.

React, Headless, and API Clients Handle It Badly

Something the usual tutorials skip: on a headless build, the maintenance lock does not show a friendly notice. WordPress returns a 503 with a Retry-After header to every request, including /wp-json/, so a React or Vue front end gets an unparseable response and typically renders a blank route or a generic fetch error.

Two habits save a lot of debugging time. First, treat a 503 from the REST API as a distinct state in your client and show a short retry message instead of a crash. Second, make static builds fail loudly rather than publishing an empty page, since a build that runs during a 20-second update window will silently ship nothing. If you run this architecture, our notes on WordPress as an app backend cover the hosting side.

Keeping Updates From Locking the Site Again

Update in batches of three to five plugins rather than selecting all of them.
Run major plugin, theme, and core updates on a staging copy first, then push.
Stay on PHP 8.1 or newer with a generous execution limit.
Take a backup before the update, not after you notice the screen.
Leave the browser tab open until the update screen reports success.

Hosts that monitor failed updates catch the orphaned file before your visitors do, which is a large part of what managed WordPress maintenance is paying for. WordPress’s own common errors documentation lists this one alongside the 500 and database connection failures for the same reason.

Frequently Asked Questions

How Can I Fix the 503 Error in WordPress?

Delete the .maintenance file from your WordPress root folder, which resolves the update-related 503 in under a minute. If the 503 persists, the cause is server-side: a PHP-FPM crash, an exhausted worker pool, or a plugin fatal error. Check the PHP error log and deactivate plugins by renaming folders until the site responds.

How to Fix Service Temporarily Unavailable?

Start by refreshing after 60 seconds, since most legitimate update locks clear on their own. If the message holds, remove the hidden maintenance file, purge your cache and CDN, then confirm your host is not performing scheduled work. Persistent failures after that usually mean PHP is timing out or running out of memory.

What Does Scheduled Maintenance Mean?

In this context it means WordPress is mid-update and has deliberately taken the site offline for roughly 2 to 10 seconds per item. Nothing was scheduled by you or your host. The wording is simply the default notice WordPress serves while it replaces plugin, theme, or core files.

How to Get a WordPress Site Out of Maintenance Mode?

Remove the .maintenance file via SFTP, your host’s file manager, or rm .maintenance over SSH, then hard refresh. If the site still shows the notice, disable any maintenance-mode plugin by renaming its folder and clear every cache layer, including object cache and edge cache.

Get Your WordPress Site Back Online

If the maintenance screen keeps coming back, or you would rather not touch files over SFTP at all, our WordPress specialists can clear the lock and find the failed update behind it. See how our 24/7 WordPress support team handles update failures, usually inside two minutes of your first message.