fedora VMs fail to upgrade: unrecognized arguments: --dist_upgrade

[eloquence]

STR:

Run sudo /etc/cron.daily/securedrop-update in dom0

Expected behavior:

Al VMs are updated

Actual behavior:

Output looks like this:

fedora-30: OK
fedora-29: ERROR (exit code 20, details in /var/log/qubes/mgmt-fedora-29.log)
sd-proxy-template: OK
sd-export-template: OK
whonix-gw-14: OK
whonix-gw-15: OK
sd-svs-disp-template: OK
fedora-30: ERROR (exit code 20, details in /var/log/qubes/mgmt-fedora-30.log)
debian-9: ERROR (exit code 20, details in /var/log/qubes/mgmt-debian-9.log)
...

The fedora logs show "dnf upgrade: error: unrecognized arguments: --dist_upgrade"

[eloquence]

(Note the "fedora-30: OK" entry at the top, followed by an error for the same VM later)

[eloquence]

Incidentally, the Debian error appears to be unrelated:

Error running 'pkg.upgrade': E: The repository 'https://deb.debian.org/debian jessie-backports 
Release' does no longer have a Release file.

[sssoleileraaa]

i see this too

[conorsch]

Rather than running dist-upgrade directly, we should be reusing the update.qubes-vm Salt SLS, maintained by the Qubes team. That state file is smart enough to handle differences between Debian/Fedora intelligently, and it ensures CVE patches, e.g. #249.