In TFN 3.15, the Log4J issue got resolved.
It has been reported that this critical Log4J fix introduced an issue.
A change in the name prevents cleaning up older log files, which can result in the disk getting full.
Because of the high severity of this issue, Barco has released a new hotfix.
This hotfix removes older logfiles AND fixes the issue of the automatic cleanup.
Installation instructions are included in the hotfix.
Barco strongly recommends installing this hotfix on existing TFN 3.15 systems.
In case it is already too late and the disk is already filled 100%, installing the hotfix will not be possible due to missing space.
To check the used space, you can log in via ssh and use the command "df -h".
The disk to check is "sda1". Once it reaches 100%, immediate action is required.
In some cases, the server itself is still able to start, but not the service.
If that is the case, the following procedure can be used to manually free up space:
First, find out in which location the disk is being filled up.
There are two specific folders that each can cause this issue.
Use the following commands to check the total used space for each of both folders:
Barco-fas:
du -sh /usr/share/barco-fas/{log.dir}
Barco-fac:
du -sh /usr/share/barco-fac/{log.dir}/var/log/barco-fac
The output will be similar to this:
Except that in case of the issue being observed, it will show numbers like "22G" and similar.
Once the folder (barco-fas / barco-fac) has been found, you can then use the command "cd" to go to the corresponding folder:
cd /usr/share/barco-fas/{log.dir}
cd /usr/share/barco-fac/{log.dir}/var/log/barco-fac
Inside this folder, there will be log files with dates in their names.
sudo rm -r *-2023-01-*
This command will remove all files from January 2023.
Typically, this should free up enough disk space to be able to install the hotfix.
If not, the command can be adapted to delete files from February, March, etc.
(*2023-02-*; *2023-03-*; etc)
In case the CMS server is not able to start, thus not allowing to access it via ssh, a reinstallation is required to restore the system.
In some cases it was observed that once the server was at 100% disk usage, even removing and successfully installing the hotfix, a mandatory service was corrupted and couldn't be repaired.
The only solution for this is a full reinstallation.
The fix will also be part of the next TFN release.
This hotfix can be found at the bottom of this knowledge article,
or on the standard Support - Drivers & firmware location. (Login required.)
