Posted: September 6, 2017 at 11:50 pm
|
We’ve recently had about four customers contact us and tell us they there was an error with their booking process, and in each case the customer has been charged, but the tickets stay unpaid. I’m at a loss to be able to solve this, but looking at the error log, there are literally hundreds of rows stating a ‘valid datetime could not be generated.’ eg. [Thu Sep 07 06:14:17.343042 2017] [:error] [pid 460] [client 108.162.226.111:17046] A valid DateTime could not be generated from “-0001-11-30 10:00:00 +1000 Australia/Sydney” because the following errors occurred: <br /> Array\n(\n [warning_count] => 0\n [warnings] => Array\n (\n )\n\n [error_count] => 1\n [errors] => Array\n (\n [0] => Unexpected data found.\n )\n\n)\n <br /> PHP version: 7.0.22-0ubuntu0.16.04.1 <br /> Stack Trace: #0 [internal function]: EventEspresso\\core\\domain\\entities\\DbSafeDateTime->__sleep()\n#1 /var/www/giantdwarf.com.au/public_html/wp-content/plugins/event-espresso-core-reg/core/EE_Session.core.php(823): serialize(Array)\n#2 /var/www/giantdwarf.com.au/public_html/wp-content/plugins/event-espresso-core-reg/core/EE_Session.core.php(761): EE_Session->_save_session_to_db()\n#3 /var/www/giantdwarf.com.au/public_html/wp-includes/class-wp-hook.php(298): EE_Session->update(”)\n#4 /var/www/giantdwarf.com.au/public_html/wp-includes/class-wp-hook.php(323): WP_Hook->apply_filters(true, Array)\n#5 /var/www/giantdwarf.com.au/public_html/wp-includes/plugin.php(453): WP_Hook->do_action(Array)\n#6 /var/www/giantdwarf.com.au/public_html/wp-includes/load.php(677): do_action(‘shutdown’)\n#7 [internal function]: shutdown_action_hook()\n#8 {main}, referer: https://www.paypal.com/checkoutnow?useraction=commit&cmd=_express-checkout&token=EC-4P225423H3545703F Looking at other threads with this issue, solutions suggested involve updating PHP, enabling ‘Encode Session Data’ and checking that base64_encode/decode is working, all of which didn’t resolve these lines appearing in the logs. Help! |
|
Further looking into this I’ve found there’s around 20 booking emails which are sitting under ‘failed to send’ in the EE list today. Not sure if that’s possibly causing or caused by the above error, but it doesn’t appear in the error logs, despite customers reporting that they have been seeing an error which says emails failed to send. |
|
I’ve just attempted to make a booking myself to see if I could reproduce the error, and run into the email error message first try: http://giantdwarf.com.au/error.jpeg I suspect the issue may be related to the event as most errors seem related to that, though I couldn’t even begin to guess why. |
Hi there, May I ask which version of PHP is installed on your webserver? |
|
|
September 20, 2017 at 11:20 pm Hi sorry for the delay, only work Thursday and Friday for this company so that combined with the time difference isn’t very conducive to keeping up with these support threads. We’re running PHP 7.0.22-0. I’ve bumped up our server’s specs again which seems to have quelled the worst of the booking issues, but we’re still seeing a lot of errors. Most notably the “A valid DateTime could not be generated from “-0001-11-30 10:00:00 +1000 Australia/Sydney”” error keep cropping up a few hundred times a day, with some providing the referring URL as the payment gateway, which has me worried we’re not catching some people’s payments. |
The error you report about the invalid DateTime is very similar to what was reported here: https://bugs.php.net/bug.php?id=72011 In the bug report two items were mentioned which may lead to a solution for you: 1) If APC or any other bytecode caching is in place on your server it may help to deactivate it |
|
|
September 22, 2017 at 12:03 am That’s interesting, and also very helpful to know where the error is originating from. Unfortunately I’ve jacked up the max_execution_time from 120 to 500 and disabled OpCache, but the error is still cropping up regularly in our logs. |
I’m curious to know if those settings are being used, did you need to reboot the server so they’d take effect? In any case the bug report mentions that this bug tends to only happen if the server is under heavy load, so the way forward on this can involve: 1) Reducing the load on the server by making some optimizations. For example right now the home page on your website takes 65 http requests. Even if you go the quick and easy route of installing a plugin like Autoptimize that number, and some server strain, could be significantly reduced. 2) Increasing the resources for the server will also help with alleviating the strain. |
|
The support post ‘Unexpected data found’ is closed to new replies.
Have a question about this support post? Create a new support post in our support forums and include a link to this existing support post so we can help you.