fpapadopou / pde Goto Github PK

If one of the commands that are executed in Docker containers hangs, the whole process will hang
and the container will never be destroyed. Need to add a time interval (if possible) to the ContainerConfig object.

The same applies to some cases when a container might get very large. There has to be a way to detect when a container's disk space exceeds a threshold and kill it and then remove it. In the current setup, 6 such containers ended up occupying 30GBs of disk space. Although they had exited, docker was unable to remove them for some reason.

The installation script is meant to be executed in a VM created with the Vagrantfile provided. The script will most likely fail right after the composer installation and before the cache cleanup due to this bug.
The system tries to access the database before it's even created and as a result the process fails.
I'll try to fix it somehow.

Create a page where admins will be able to access log files from the browser. They can now only access them via ssh. There's a dummy action for that in the AdminActionController.

In order to do this, the following are necessary:

• A solid installation script
• Fixtures
• Some testing scenarios
• Unit and/or functional tests
• Travis install and test execution files

Some handler methods return assoc arrays which are not handled properly by the caller methods.
A typical example is the deleteDirectory call in TeamManager::deleteTeam where the call to deleteDirectory is placed in try-catch block, although the method never throws an exception.

Need to further investigate and detect more of these issues and resolve them properly.

An administrator should be able to delete teams and/or users. They should also be able to
backup deleted teams' projects.