Plugins Plugin hub Backup SEO Caching Security Forms WooCommerce Images Best backup plugins
Hosting Hosting directory Shared Managed Cloud India
Themes & builders Compare Rank Math vs Yoast
Tools All tools Hosting finder Plugin stack builder Cost calculator Salt generator Theme detector Who is hosting this? Speed test Headline analyzer robots.txt generator Redirect generator Uptime calculator Blog name generator
Best lists Best backup plugins Deals
Guides All guides Speed up WordPress Choose hosting Back up WordPress
How we test

Troubleshooting · Troubleshooting

How to Fix cURL Error 28: Connection Timed Out in WordPress

Short answer

cURL error 28 means a WordPress HTTP request ran out of time before it got a reply. First, open Tools, then Site Health, and copy the exact message, including the destination and the number of milliseconds. Then fix the cause that matches that destination: a blocked loopback, a blocked WordPress.org connection, a slow plugin service, or a DNS or firewall fault.

Key takeaways

  • cURL error 28 is a timeout, not proof that WordPress code is broken.
  • WordPress gives most HTTP requests 5 seconds by default, while Site Health gives its WordPress.org test 10 seconds.
  • A loopback timeout points at your own site, while a WordPress.org timeout points at outbound traffic.
  • Raising a timeout hides the fault when the real cause is a firewall, DNS fault or slow service.
  • If the same error returns after plugin and firewall checks, the host must check server networking.

What does cURL error 28 mean in WordPress?

cURL error 28 means a WordPress HTTP request reached its time limit before it got a reply.

A request is a message WordPress sends to another server and waits to hear back from. cURL is the library PHP uses to send that message. When the wait runs out, cURL stops the request and reports error 28.

The libcurl error list names code 28 CURLE_OPERATION_TIMEDOUT and defines it as an operation timeout, where the set time limit was reached. That wording matters because it tells you the request timed out, not that the remote server sent an error page.

WordPress sends most of these requests through its HTTP API, using functions such as wp_remote_get(). The WordPress HTTP API handbook lists a default timeout of 5 seconds for wp_remote_get(), so a slow reply can fail even when the remote service is still working.

Where does the error show up first?

Open Tools, then Site Health, in your WordPress admin to find the exact failed test.

Site Health runs separate checks, so the same error number can come from different places. Copy the full message, the destination shown, and the number of milliseconds before you change anything.

Common places for the message include:

  • Site Health, under “Could not reach WordPress.org” or a failed loopback test
  • Dashboard, then Updates, when a core, theme or plugin update check fails
  • A plugin log, when a licence check, payment callback or feed import times out
  • The block editor, when a REST API request fails while saving or loading

The destination tells you where to look next. A request to api.wordpress.org is an outbound request, while a request to your own domain is a loopback request. Treat those two cases as different faults.

What causes cURL error 28 most often?

The most likely causes are a slow remote service, a blocked loopback, a firewall rule, or a DNS fault.

Start with the destination and the timeout value, then work through the causes in this order. The table below links each sign to the cause it points at.

SignMost likely causeWhere to check first
Timeout after 10000 milliseconds on your own domainBlocked loopback requestTools, then Site Health, then loopback test
Timeout after 10000 milliseconds to api.wordpress.orgBlocked outbound trafficHost firewall and DNS settings
Timeout after 5000 milliseconds to one plugin serviceSlow or down remote serviceThat plugin log and its service status
“Resolving timed out” in the messageDNS faultHost DNS settings
Error appears after a new plugin or firewall changePlugin or firewall blockPlugin list and firewall rules

Kinsta lists the same main causes in its cURL error 28 guide: a firewall that treats a REST API request as suspicious, a faulty DNS server, a low server timeout limit, and a plugin that disrupts the cURL transfer.

A firewall can block the request without sending a reply, so cURL waits until the clock runs out. DNS faults have the same effect when the server cannot turn a hostname into an address in time. A slow service is different again: the connection opens, but the reply takes longer than the limit.

How do you read the timeout number in the message?

The number of milliseconds shows how long cURL waited, and the words around it show which stage failed.

Read the full line, not just the number 28. The common forms are:

  • Operation timed out after 10000 milliseconds with 0 bytes received means no reply arrived in 10 seconds.
  • Connection timed out after 10001 milliseconds means the connection itself never opened.
  • Resolving timed out after 5001 milliseconds means DNS lookup did not finish in time.

The 10000 millisecond value is a strong clue. Site Health uses a 10 second timeout for its WordPress.org test, as shown in the WordPress.org communication test.

Its loopback test uses the same 10 second timeout, as shown in the loopback test. A 5000 millisecond value points at the 5 second default for a normal WordPress HTTP request.

Note the exact value and destination in a short note before you start fixing. If the value changes after a fix, that change tells you whether the fix helped. If it stays the same, the cause is still present.

How do you fix a loopback timeout?

A loopback timeout means WordPress cannot call its own site, so check plugins, security rules and the server path to your own domain first.

A loopback is WordPress connecting back to itself. WordPress uses loopbacks to run WP-Cron, publish scheduled posts, and check theme and plugin edits. The loopback test posts to wp-cron.php with a 10 second timeout, according to the loopback test reference.

Follow these steps in order:

  1. Go to Tools, then Site Health, then Status, and open the loopback item to confirm it is the failed test.
  2. Install the Health Check plugin from Plugins, then Add New, and open its Troubleshooting Mode, which disables plugins only for your logged-in session.
  3. Re-run the Site Health loopback test while Troubleshooting Mode is on.
  4. If the test passes, turn plugins back on one at a time and re-run the test after each one.
  5. If the test still fails with all plugins off, check your security plugin and any firewall for rules that block wp-cron.php, admin-ajax.php or /wp-json/.

The Make WordPress Support handbook explains that Troubleshooting Mode gives your logged-in user a plain version of WordPress with plugins off and a default theme, while visitors still see the normal site. That makes it safe to use on a live site for this check.

If your site sits behind a proxy or a firewall service, that service can block the loopback before it reaches WordPress. Ask your host whether the server can reach your own domain by name.

How do you fix a WordPress.org timeout?

A WordPress.org timeout means outbound traffic is blocked or slow, so check DNS, the firewall and the server route to api.wordpress.org.

Site Health tests this with a request to https://api.wordpress.org and a 10 second timeout. WordPress uses that link to check for new versions and to install and update core, themes and plugins, as the WordPress.org communication test states.

Follow these steps in order:

  1. Copy the exact Site Health message, including the IP address shown for WordPress.org.
  2. Ask your host to run a DNS lookup for api.wordpress.org from the server and to fetch https://api.wordpress.org with a 10 second limit.
  3. Ask the host to confirm that outbound traffic on TCP ports 80 and 443 is open.
  4. Ask the host to check whether a firewall, NAT rule or security filter drops outbound requests without a reply.

Kinsta states that a WordPress firewall can treat a REST API request as suspicious and block it, and that a faulty DNS server or a low server timeout limit can also cause cURL error 28. That guidance is in the Kinsta cURL error 28 guide.

Do not turn off SSL checks to fix this. WordPress checks SSL by default, and its HTTP API uses its own certificate bundle at wp-includes/certificates/ca-bundle.crt. Turning off that check would hide a certificate fault and weaken security.

How do you fix a slow plugin or remote service timeout?

When only one plugin or service times out, the fault is usually that service, so isolate the plugin before you change server settings.

A single slow service can stall the admin screen or the page that calls it. WordPress waits up to 5 seconds by default for each request, so several slow calls in a row can add up to a long delay.

Follow these steps:

  1. Note which plugin shows the error and which address it calls.
  2. Turn that plugin off on a staging copy, or use Troubleshooting Mode, and repeat the same action.
  3. If the error stops, update the plugin, check its API key and settings, and contact its vendor with the exact error, time and destination.
  4. If the error continues with the plugin off, the fault is in the server path, not that plugin.

For code you control, PHP gives two separate cURL limits. The PHP curl_setopt manual documents CURLOPT_CONNECTTIMEOUT for the connection stage and CURLOPT_TIMEOUT for the whole transfer.

WordPress exposes the same idea through the timeout argument in wp_remote_get(), which the wp_remote_get reference lists with a default of 5 seconds.

Raise a timeout only when the remote task truly needs more time and the connection is otherwise healthy. Raising it for a blocked or dead service only makes the page wait longer before it fails.

How do you check the fix worked?

Re-run the exact Site Health test or plugin action that first showed the error, and check that scheduled tasks run again.

A fix is confirmed only by the same test that failed. Do not rely on the front of the site loading, because a site can serve visitors while its outbound or loopback requests still fail.

Check these items:

  • Tools, then Site Health, then Status, shows the failed test as passed.
  • Dashboard, then Updates, loads the update check without a cURL error.
  • The plugin action that timed out now completes, such as a licence check or import.
  • A scheduled post publishes at its set time, which shows WP-Cron is running.

If you use a staging copy, repeat the same test there after each change. Keep a short note of the test, the time and the result, so you can show your host what changed.

When should you contact your host?

Contact your host when the error stays after plugin and firewall checks, or when every outbound request fails.

Your host controls the parts you cannot see from the admin screen: DNS, the firewall, NAT rules, the cURL version and server load. Give the host the exact message, the destination, the timeout value, and the time it happened.

Ask the host to check:

  • DNS lookup for the destination from the server
  • Outbound HTTPS to the destination with a 10 second limit
  • Firewall and NAT rules for outbound traffic and loopback traffic
  • The installed cURL version and whether the PHP cURL extension is present

The WordPress Hosting Handbook lists cURL as a highly recommended PHP extension, and notes that PHP 8.0 needs libcurl 7.29.0 or newer, while PHP 7.3 needs libcurl 7.15.5 or newer. That detail is in the Hosting Handbook server environment page.

WordPress.org recommends PHP 8.3 or greater, MariaDB 10.11 or greater or MySQL 8.4 or greater, and HTTPS support, as of November 2026. Those figures are on the WordPress.org requirements page.

An old PHP or cURL version is not the usual cause of error 28, but an outdated server can still add delay. Include your PHP and cURL versions in the message to the host.

A related timeout fault is a gateway timeout, which has its own causes and fixes. Read the guide on how to fix the 504 gateway timeout if your site shows that error instead.

For a slow site in general, read how to tell if your host is slowing WordPress down.

What mistakes should you avoid?

Do not raise timeouts, turn off SSL checks, or edit server files at random to make the message go away.

These mistakes are common:

  • Raising the timeout to a large value when the request is blocked, which only delays the same failure.
  • Turning off SSL verification, which hides certificate faults and weakens security.
  • Deactivating every plugin on the live site at once, instead of using Troubleshooting Mode or a staging copy.
  • Changing DNS, firewall or PHP settings without noting the old values first.
  • Treating one failed plugin as proof the whole server is broken, when only that service is slow.

Fix one cause at a time and re-run the same test after each change. That way you know which change worked. If you change several things at once, you cannot tell which one fixed the fault, and the error may return later without a clear cause.

Before you change PHP settings, check the current version in Tools, then Site Health, then Info. Read how to update PHP for WordPress before you change the PHP version.

Also check the WordPress hosting requirements to confirm the versions your host should run.

Frequently asked questions

Is cURL error 28 caused by WordPress itself?

No, not usually. The error means a request timed out, and the cause is often a slow remote service, a firewall rule, a DNS fault or a blocked loopback. WordPress reports the error because it made the request, but the fault is often in the network path or the remote service. Check the destination in the message first, because that points at the cause.

How long does WordPress wait before cURL error 28 appears?

It depends on the request. Most WordPress HTTP requests use a 5 second default timeout. Site Health uses a 10 second timeout for its WordPress.org test and its loopback test. The message shows the exact wait in milliseconds, such as 5000 or 10000, so read that number to see which limit was hit.

Can a plugin cause cURL error 28?

Yes. A plugin can make a slow or blocked request to its own service, and that request can time out. A security plugin can also block a loopback or REST API request. Use Troubleshooting Mode to test with plugins off for your session only, then turn plugins back on one at a time to find the one that brings the error back.

Should I increase the timeout to fix cURL error 28?

Only when the remote task truly needs more time and the connection is otherwise healthy. If a firewall blocks the request, DNS fails, or the service is down, a longer timeout only delays the same failure. Fix the block or the slow service first, then raise the timeout only for that specific request if it still needs more time.

Why does my site load fine but Site Health still shows cURL error 28?

Serving pages to visitors and making outbound or loopback requests are different jobs. A site can serve cached pages while its server cannot reach WordPress.org or itself. That is why Site Health can fail while the front of the site looks normal. The failed test matters because updates, scheduled posts and plugin services depend on those requests.

Sources

  1. wp_remote_get() WordPress Developer Resources
  2. WP_Site_Health::get_test_dotorg_communication() WordPress Developer Resources
  3. WP_Site_Health::can_perform_loopback() WordPress Developer Resources
  4. curl_setopt - PHP Manual PHP.net
  5. Requirements WordPress.org
  6. Server Environment Make WordPress Hosting
  7. How To Fix the cURL Error 28: Connection Timed Out Kinsta

Is your product missing, or something out of date?

Tell us what to add or correct. We read every note and reply when we need a source. A message does not buy a ranking.

About the editors

WPLegible editors

Researched from the primary sources listed above and checked against our editorial rules. About the site · How we test