agssl

I just received the response. Hetzner no longer uses mod_php (at least not from PHP 8 onwards). The setting simply needs to be added without the “IfModule mod_phpX.c” directive ??

Just write it like

# BEGIN NinjaFirewall
php_value auto_prepend_file "[…]/wp-content/nfwlog/ninjafirewall.php"
# END NinjaFirewall

I created a ticket and just received a response that Hetzner can manually set this in the PHP.ini for each account. I also asked if there’s any way to make the change myself, but haven’t received an answer on that yet. Creating a support ticket with Hetzner for each site is a bit tedious. I’ll post an update here once Hetzner gets back to me.

Hey, I just did that. Here’s the relevant part (I masked the paths with […]):

HTTP server	:	Apache
PHP version	:	8.2.23
PHP SAPI	:	CGI-FCGI

auto_prepend_file	:	none
.htninja	:	found in […]/.htninja
wp-config.php	:	found in […]/wp-config.php
NinjaFirewall detection	:	NinjaFirewall WP Edition is loaded (WordPress WAF mode)

Loaded INI file	:	/home/httpd/php82-ini/[…]/php.ini
user_ini.filename	:	none
user_ini.cache_ttl	:	300 seconds
User PHP INI	:	php.ini found -
 	 	 
DOCUMENT_ROOT	:	[…]
ABSPATH	:	[…]
WordPress version	:	6.6.2
WP_CONTENT_DIR	: […]/wp-content
Plugins directory	:	[…]/wp-content/plugins
User Role	:	Administrator
User Capabilities	:	manage_options: OK - unfiltered_html: OK
Log dir permissions	:	[…]/wp-content/nfwlog dir is writable
Cache dir permissions	:	[…]/wp-content/nfwlog/cache dir is writable

Hi,

I’ve just opened a ticket.

Thanks!

Hi,

yes I’m using 2.3.3. The behavior can be reproduced both in the local test environment and on a live server.

Thanks!

Hey,

thanks for the update. Unfortunately it’s still not working. Neither with {set_password_url} nor with {reset_password_url}. The lines in then final mail are just empty.

Here’s a screenshot of the mail editor

and here’s a second screenshot with the outcoming mail

Thanks in advance!

Great, thanks. I’ll wait for your reply.

Hey Tom,
really strange. It doesn’t matter if the theme.json is empty or if there’s anything inside it like:

{
    "version": 1,
    "settings": {
        "layout": {
            "contentSize": "840px",
            "wideSize": "1100px"
        }
    }
}

The align/wide option of the container block is missing in any case. I don’t need the theme.json right now, but was playing with it purely out of interest – I just thought it would be helpful for you guys.

Hi David,

yes, the alignment option is available for other blocks – like the wp core blocks image, cover, video or table. It also doesn’t affect my custom blocks (acf). As far as I can see, it only affects the container block right now – but of course I haven’t tested all blocks and can’t guarantee that it’s still missing in another block.

Thanks Leo,

I’ll definitely give it a try and additionally use the presales form to clarify some aspects of my request.

The problem is solved. It was related to IPv6 connectivity or better the faulty
connectivity.

Indeed, the original error must have been caused indirectly in some way. The main problem seems to be the API calls. I will ask the hoster again, what could cause the problem.

Thanks for your fast reply!

The site is running on PHP 7.2.24-0ubuntu0.18.04.7 with the latest WordPress Version 5.5.1

I already did a reset, but it didn’t help. Only .forceOff or renaming the Plugin will bring the site back. With the activated Shield Plugin the site keeps loading forever – without, the site is super fast.

Memory limit is set to 256MB and the free disk space is > 3GB
Zend Opcache is up and running, but I also did a test with disabled Opcache, with the same result.

I’ve just duplicated the site to an staging environment and deactivated all plugins except of Shield. Still the same result. The site keeps loading for minutes (but loads afterwards) – it’s just not usable.

I don’t get any errors in the console and the server is bored (actually it is oversized for this site). The debug.log is also empty this time.

Update: I’ve installed Query Monitor and get some API-Call errors:
unable to connect to ssl://wphashes.com:443 (Connection timed out)

Here’s a screenshot:
https://ibb.co/yS216BM

Awesome, thanks for the hint Beau. I tried your code, but got an internal error – a slightly modified version works like a charm. So, if anyone else needs it:

add_action( 'keyring_post_imported', function( $post_id ) {
  $post = array(
    'ID'    => $post_id,
    'post_status' => 'draft',
  );
  wp_update_post( $post );
} );

This would be great thanks.

By the way … I found an ugly workaround for the moment. You can schedule an update in the second language, but the scheduled duplicate will appear on the page of the first language. After the release date, everything is in place again and the second language page has been replaced.