date.timezone Fatal after PHP update 7.9.7

šŸ’¬ General Discussion
1

Recently needed to update to 7.9.7 because workflows was not working in 7.9.6. This required me to updat PHP which is now Apache/2.4.23 (Fedora) PHP/7.1.8

I am still getting a error every minuteā€¦

Mon Nov 13 17:27:01 2017 [20389][-none-][FATAL] Configuration variable date.timezone is not set, guessed timezone UTC. Please set date.timezone=ā€œUTCā€ in php.ini!

I have updated the php.ini in the root of the suite install to date.timezone= ā€œAmerica/New_Yorkā€ I have updated the CLI in /etc/php7/php.ini as well to the same. I tried it with both ā€œā€ and without.

I have varified my date of Server and it is correct. I have verified the user date and it is correct. I have verified the Phpinfo () and it is correct showing

date/time support enabled
timelib version 2017.05beta9
ā€œOlsonā€ Timezone Database Version 2017.3
Timezone Database internal
Default timezone America/New_York
Directive Local Value Master Value
date.default_latitude 31.7667 31.7667
date.default_longitude 35.2333 35.2333
date.sunrise_zenith 90.583333 90.583333
date.sunset_zenith 90.583333 90.583333
date.timezone America/New_York America/New_York

Any thoughts on trying to troubleshoot resolving the date.timezone fatal error?

2

The problem is that cron is running from a different PHP (called CLI), and thus from a different php.ini.

You can try adding that setting to the cron job definition by passing it from the command-line into PHP (-d option)

Something like this:


*    *    *    *    *     cd /path/to/suitecrm; /path/to/php-cli/php -f cron.php -d date.timezone="Europe/Lisbon" > /dev/null 2>&1

You can ask your hosting for details (correct paths) or for alternatives, if this does not work. It is their responsibility to provide you with a way to correctly set options for CLI PHP.

3

Or find out the location of your php.ini for cli by entering at the command prompt

php --ini

and change accordingly

4

Makes sense but I need to clarify one item.

As originally stated I have updated the CLI PHP.ini but was still getting the error.

Do I place this cron directive in the same location as the scheduler cron is running? Will this new cron address the issue even though originally I had changed the date.timezone in the CLI which for me is found in the /etc/php7/php.ini? or is this overiding it with the cron running instead of the php.ini directive?

5

OK so I think I have this correct for my main database but it throwing errors for my dev site on the same server.

Any chance you see something that might be causing this?

Warning: Declaration of SugarDateTime::setTime($hour, $minute, $sec = 0) should be compatible with DateTime::setTime($hour, $minute, $second = NULL, $microseconds = NULL) in /var/www/sites/suitecrm_dev/rp/include/SugarDateTime.php on line 692

Warning: Declaration of SugarEmailAddress::save($id, $module, $new_addrs = Array, $primary = ā€˜ā€™, $replyTo = ā€˜ā€™, $invalid = ā€˜ā€™, $optOut = ā€˜ā€™, $in_workflow = false) should be compatible with SugarBean::save($check_notify = false) in /var/www/sites/suitecrm_dev/rp/include/SugarEmailAddress/SugarEmailAddress.php on line 1162

Warning: session_start(): Cannot start session when headers already sent in /var/www/sites/suitecrm_dev/rp/include/MVC/SugarApplication.php on line 619

Warning: Cannot modify header information - headers already sent by (output started at /var/www/sites/suitecrm_dev/rp/include/SugarDateTime.php:692) in /var/www/sites/suitecrm_dev/rp/include/utils.php on line 2802

Warning: session_destroy(): Trying to destroy uninitialized session in /var/www/sites/suitecrm_dev/rp/include/MVC/SugarApplication.php on line 124

It seems to be a date configuration issue.

6

Hi, I think your error_reporting is set incorrect. Unfortunately the code of SuiteCRM still needs some clean up to get rid of all the warnings. Please check your error_reporting in php.ini and try the following value:

error_reporting = E_ALL & ~E_DEPRECATED & ~E_STRICT & ~E_NOTICE & ~E_WARNING

Restart your nginx/apache/php_fpm after changing the value.

7

Turned on error reporting error_reporting = E_ALL & ~E_DEPRECATED & ~E_STRICT & ~E_NOTICE & ~E_WARNING
In the /etc/php7/php.ini file and the error_log is showing

[proxy_fcgi:error] [pid 937] [client ip:6987] AH01079: failed to make connection to backend: httpd-UDS
[Wed Nov 15 08:36:31.904435 2017] [proxy:error] [pid 960] (2)No such file or directory: AH02454: FCGI: attempt to connect to Unix domain socket /run/php-fpm/www.sock (*) failed

I suspect it is because php-fpm does not auto restart but I donā€™t know how to check or restart. Thoughts?

8

Hi, list your services by

service --status-all

See if you can find a service by the name of php-fpm or php7.1-fpm. Then

service php-fpm restart

Also check your logs, like

/var/log/php-fpm.log
9

services only show configured devices
lo eth0

I have attempted restart of php-fpm

php-fpm.log

{15-Nov-2017 09:26:28] ERROR: FPM initialization failed
[15-Nov-2017 13:48:58] ERROR: unable to bind listening socket for address ā€˜/run/php-fpm/www.sockā€™: No such file or directory (2)
[15-Nov-2017 13:48:58] ERROR: FPM initialization failed

10

@tseward I never installed using php-fpm, but maybe you can look into which modules youā€™re loading. You might have required modules missing, and that could be the cause of these errors.

On Ubuntu 16.04 with PHP 7.0 I typically use these

apt install zip unzip php-mbstring php7.0-mbstring php-gettext php-xml-parser php7.0-zip php7.0-imap php7.0-gd php7.0-curl sendmail phpmyadmin

You probably need to adapt that for PHP 7.1, but it can serve as a guide.

11

Hi,

this is a system configuration issue.

Please check if the directory /run/php-fpm exists, if not create it and make sure the system user that runs php-fpm (usually ā€œwww-dataā€ or ā€œwwwā€) can write in this directory.