Out of space issue

Disk space shortage was solved (at the end of January 2020) for the Blast databases by addition
of 1 TB to both the Galaxy production and test servers. Though this solved the issue for these
databases, there still is the issue that Galaxy's temporary database (/home/galaxy/galaxy/database/tmp)
fills up (causing /dev/sda1 to run out of its 58GB of diskspace).

To check diskspace usage:
df -h

Obviously this should be taken care of with cron jobs but there are still issues with
getting these to work in the current version of Galaxy. So for the time being, just manually empty
this temporary database:

rm -r /home/galaxy/galaxy/database/tmp/*

/dev/sda1

Actually most of /dev/sda1 seems to be taken up by superseded conda packages (/opt/anaconda3/pkgs).
Remove unused packaged with:

conda clean -a

Other folders (at least on Galaxy-Test) to keep an eye on are:
/opt/anaconda3/envs and /home/galaxy/Tools

/dev/sda1 with Ansible

If galaxy was installed using Ansible /dev/sda can run out of space because of automated backups of the
PSQL database. To quickly regain diskspace remove all but the last backup (keep current!) from /data/backups/

clear journalctl

To keep the last two days only:
sudo journalctl --vacuum-time=2d
To keep the only the last 100 MB of log files:
sudo journalctl --vacuum-size=100M