I guessed the same thing.
That's why I was surprised that the nginx reload and php-fpm restart commands that I copied and pasted from the docs failed to solve the problem.
I also tried to clear Symfony's cache via the command line: "php symfony cache:clear". But that just generated an error message about no cache present. I still think that this smells like some sort of stale cache problem, but since I couldn't figure out how to unstick whatever was stuck, I settled for the reboot option.
The problem is easily reproducible:
Start by booting the Ubuntu ISO image (I used ubuntu-18.04.5-desktop-amd64.iso).
sudo apt update, and sudo apt install build-essential dkms linux-headers-$(uname -r).
After that, follow the AtoM 2.6 Ubuntu 18.04 LTS installation instructions.
In the MySQL wget step, replace "mysql-apt-config_0.8.15-1_all.deb" with the name of the latest MySQLversion.
Skip the optional "sudo mysql_secure_installation" step.
Skip the optional "
If you are not planning to use the default PHP pool (www), feel free to
remove it" step.
For AtoM, choose "Option 1: Download the tarball".
Perform all other recommended steps.
By the way, even after the reboot "fix", the following message still appears in qubit_worker.log. I don't know if it's related to the installation problem, but in any case, I'm ignoring it for now:
symfony [crit] {PropelException} Unable to open PDO connection [wrapped: SQLSTATE[HY000] [2002] No such file or directory]
Are there any logging or debugging options that could be helpful in tracing execution of the installation steps?
-Scott