Re: Better communication and co-ordination for GVM (OpenVAS)
Hi,
Thanks for your email.
I'm currently the main uploader fo GVM packages with Raphaël.
Le 20/01/2021 à 11:53, Björn Ricks a écrit :
[...]
With this email I would like to ask how the current situation can be
improved and how we can reduce the frustration on both sides (upstream
and downstream). For example currently a lot o people are popping up at
our forum (https://community.greenbone.net/) having issues with the
setup on Kali. For us it is very difficult to help in this case and it
would be nice if some Kali developers could join the forum and assist
users with their setup.
I really can't answer to all the questions/issues that people have about GVM setup/installation on Kali or Debian.
I try to help and answer when people open issues on bugs.kali.org (or bugs.debian.org) especially after an update.
The /usr/share/doc/gvm/README.Debian gives information to the user about the GVM setup and installation.
As you may know, Debian/Kali provides a gvm-setup script. It automates the setup like downloads, database creation, following the instructions found in the INSTALL.md files of the different GVM parts. We really tried to modify as little as possible the installation.
The main difference is that we install the packages in / and not a specific directory, we created a special user _gvm to have access to directories like /var/log/gvm /var/run/gvm (the "normal" user doesn't have access to such directories).
Maybe it can help the packages maintainers in the different distributions if you provide a recommended way to install in a system wide manner (FHS compliant).
I also think that one of the biggest issue users have is the postgresql version:
in Debian we compile gvmd with the packaged Postgresql version in Debian/Unstable.
And gvmd required that the Postgresql cluster main on port 5432 must have the same version as the one used to compile the package.
But in Debian the Postgresql cluster main on port 5432 is on the first Postgresql version you have installed and remains like this if the user doesn't do anything.
Users need to use pg_upgradecluster to upgrade the cluster. We don't want to automate this in the setup because it may break things for the user. But a lot of people don't know about this.
I don't know if there is something you can do to use the main cluster even if it's not the Postgresql version used for compilation of gvmd.
Best regards,
Sophie
Best regards
Björn
P.S.: I am not subscribed to the list. So please keep me in CC when
answering.
Reply to: