PHP-FPM 7.3 on Ubuntu 18.04

ZipperZapper

Verified User
Joined
Nov 30, 2015
Messages
142
As mentioned in the Debian 10 Alpha topic, I'm in the process of installing DirectAdmin on Ubuntu 18.04 LTS, based on Debian 10. To keep that topic clean I opted for a fresh topic for the problem I haven't found a solution for yet.

As summed up in this post, I managed to install DirectAdmin on Ubuntu 18.04 after switching my license to Linux 64-bit, as it would fail when set to Debian 10 Beta.

Most of it runs fine now, after setting systemd=1 in the DirectAdmin conf. One bug is still there and might be very specific to Ubuntu, I don't know. As I bought my license through my hoster I don't have access to the Ticket area of DA to open one so I hope people are able to debug it from here.

As mentioned in the other topic, PHP-FPM 7.3 fails on Ubuntu 18.04, because some library seems to be missing:

Code:
php-fpm73.service - The PHP FastCGI Process Manager
Loaded: loaded (/etc/systemd/system/php-fpm73.service; enabled; vendor preset: enabled)
Active: failed (Result: exit-code) since Sun 2019-11-03 16:21:10 CET; 4s ago
Process: 900 ExecStart=/usr/local/php73/sbin/php-fpm73 --nodaemonize --pid=/run/php-fpm73.pid (code=exited, status=127)
Main PID: 900 (code=exited, status=127)

systemd[1]: Starting The PHP FastCGI Process Manager...
php-fpm73[900]: /usr/local/php73/sbin/php-fpm73: error while loading shared libraries: libicudata.so.64: cannot open shared object file: No such file or directory
systemd[1]: php-fpm73.service: Main process exited, code=exited, status=127/n/a
systemd[1]: php-fpm73.service: Failed with result 'exit-code'.
systemd[1]: Failed to start The PHP FastCGI Process Manager.

This is 'fixed' by removing the following from the PHP-configure file, but that's just a temporary workaround.

Code:
        --with-icu-dir=/usr/local/icu \
        --enable-intl
 
Please try adding this line to /etc/ld.so.conf:
/usr/local/icu/lib

Then run:
ldconfig
 
Wow, that worked! Thank you very much.

After some more testing I find the running of services is a bit buggy. The DirectAdmin service won't start after every reboot, but will start without issues with a 'sudo service directadmin start'.

Other services like nginx won't start automaticly most of the time either. Setting systemd=1 or systemd=0 both produce different errors when using the Service Monitor start/stop/restart buttons in the DA UI:

With systemd=0:
Code:
Cannot find the script

Details
/etc/init.d/nginx

With systemd=1
Code:
An error has occurred

Details
/usr/bin/systemctl start nginx.service 2>&1
 
No errors. It starts fine with 'sudo service nginx start' too, it just won't start on a reboot. Same goes for directadmin service.

The Service Monitor gives the same errors for every service.
 
Both DirectAdmin and Nginx not starting at boot was fixed by setting a static IP which I forgot at first...
Apparently a slight delay in DHCP-assigning causes both services to fail at first.
 
Back
Top