stanley@litespeed

Seems your hosting provider disabled the fsockopen function or firewalled outbound requests, please contact them about this question with your testing script to confirm the function is enabled.

Could you try the following script for testing the ping function?

<?php

$domain = 'us.wp.api.litespeedtech.com' ;
$starttime	= microtime( true ) ;
$file		= fsockopen( $domain, 80, $errno, $errstr, 10 ) ;
$stoptime	= microtime( true ) ;
$status		= 0 ;

if ( ! $file ) {
	echo "---Ping error–--" ;
	echo "\r\n" ;
	echo $errno ;
	echo "\r\n" ;
	echo $errstr ;
} else {
	fclose( $file ) ;
	$status = ( $stoptime - $starttime ) * 1000 ;
	$status = floor( $status ) ;
	print_r( $status );
}

Cloudflare API “Purge Files by Cache-Tags or Host” just support Enterprise user, so in general, our plugin can not just purge a subdomain.

May I know are you using the Enterprise version of Cloudflare and wish to have this function?

Hi @audiomonk, the setting does not require any IP whitelist. Have you checked that is the DNS (DNS lookup tools) already updated? If yes, could you provide you DNS setting with us for finding out the problem? (Please hidden the root domain before reply)

Hi @mikehermary, usually flash of unstyled content is caused by “Load CSS Asynchronously”, you need to turn ON “Generate Critical CSS” to prevent the FOUC.

I cannot find the FOUC issue on your site now, please check after you turned “Generate Critical CSS” ON, you can find <style id=”litespeed-optm-css-rules”>.

If you the inner text of this style is empty, that means CCSS is not generated yet. That’s why you still can meet unstyled content when your internet speed is not fast enough. Then you can turn off “Generate Critical CSS In Background” to force it generated when visit.

Hi @horrormoviesgr, can you try to make a simple PHP curl request like:

<?php
$ch = curl_init();
curl_setopt($ch, CURLOPT_URL,"https://eu.wp.api.litespeedtech.com");
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
$server_output = curl_exec($ch);
curl_close ($ch);
print_r($server_output);

to see does it also show timeout?

Hi @mirishehu, may I know what is the full URL you put on it? (You can hide the root domain)

Hi @scmsteve thanks for the details report, this issue might be related to our recently found bug.

Could you try to comment out rows 453 – 455 in plugin/litespeed-cache/inc/object.class.php:

if ( ! $this->_can_cache() ) {
	return null ;
}

and run the add option again to see does it back to normal?

Hi @menathor, please put the root domain (myotherdomain.com) on Cloudflare API > Domain and you are correct that if putting with subdomain name might confusing the plugin on the URL search.

May I know how can you get into the dashboard if all the links are showing the error?

An update for anyone experiencing this issue.

If you click “Initial optimization” with no reponse, please check on the debug.log.

`
[IAPI] posting to : https://eu.wp.api.litespeedtech.com/media_sync_data
[IAPI] failed to post: cURL error 28: Connection timed out after 10001 milliseconds
It’s connection timeout when calling to our optimization server, please try to create a simple post request to https://eu.wp.api.litespeedtech.com/media_sync_data
`

If you find these two rows that mean the connection between IMG OPTM server and your server is failed. Please confirm you already added IPs on all of your server firewall.

Next steps please try to make a curl request to the your log “posting to” address, for example:

<?php
$ch = curl_init();
curl_setopt($ch, CURLOPT_URL,"https://eu.wp.api.litespeedtech.com/media_sync_data");
curl_setopt($ch, CURLOPT_POST, 1);
curl_setopt($ch, CURLOPT_POSTFIELDS, "postvar1=value1");
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
$server_output = curl_exec($ch);
curl_close ($ch);
print_r($server_output);
?>

If it’s still timeout please try to change the curl URL to other address, e.g. us.wp.api.... to confirm it’s just an issue with a single server.

If the connection timeout is just from the first URL, our server might be on heavy load or your IP has been blocked due to some reason. You can wait for a while or remove the litespeed_api_cloud on database options table for change the server location. (Just for the experienced user, this action might break the site if not doing correctly.)

Hi @julianmarin, may I know how could you deactivate the LiteSpeed Cache plugin? By WP CLI or you have any method to bypass the error?

• This reply was modified 5 years, 1 month ago by stanley@litespeed.

The remaining browser cache issue is from an external source (as the following list), so your site cannot control their browser cache.

https://www.googletagmanager.com/gtag/js?id=UA-123098187-1 (15 minutes)
https://connect.facebook.net/en_US/fbevents.js (20 minutes)
https://connect.facebook.net/signals/config/1070268079839004?v=2.9.15&r=stable (20 minutes)
https://connect.facebook.net/signals/plugins/identity.js?v=2.9.15 (20 minutes)
https://www.google-analytics.com/analytics.js (2 hours)

There’s some suggestion on Stackoverflow for your reference.

Your welcome @createpop. Happy to read you solve your issue.

Hi @lapomme, may I confirm that you are already followed these steps to request a webp version from image optimization? If yes, could you check on the media page and find are those “WEBP” icon dimmed (greyed out)?