Errors Returned when Updating to Fedora-32

Would somone provide help with evaluating errors returned during an upgrade from Fedora-30 to Fedora-32?

It takes awhile for the dnf command to complete so I was hoping for help. The last two look as if I may not have a repository configured correctly, but that’s a wild guess.


Using “Upgrading Fedora TemplateVMs” as a reference (1), running the following command:

[user@fedora-32 ~]$ sudo dnf --releasever=32 --setopt=cachedir=/mnt/removable --best --allowerasing distro-sync

returned the following errors: (For brevity, ellipses (…) designate skipped output)

Running transaction

Running scriptlet: qubes-kernel-vm-support-4.0.30-1.fc30.x86_64 1493/2899
Error! DKMS: Remove cancelled because --rpm_safe_upgrade scenario detected.
Cleanup : qubes-kernel-vm-support-4.0.30-1.fc30.x86_64 1493/2899

Cleanup : NetworkManager-1:1.16.4-1.fc30.x86_64 2075/2899
Running scriptlet: NetworkManager-1:1.16.4-1.fc30.x86_64 2075/2899
Failed to wait for daemon to reply: Broken pipe

Cleanup : glibc-2.29-29.fc30.x86_64 2886/2899
Running scriptlet: glibc-2.29-29.fc30.x86_64 2886/2899
/var/tmp/rpm-tmp.IT2iKC: line 5: /sbin/sln: No such file or directory

Running scriptlet: qubes-kernel-vm-support-4.0.30-1.fc32.x86_64 2899/2899
dracut: Cannot find module directory /lib/modules/*/
dracut: and --no-kernel was not specified

Errors during downloading metadata for repository ‘fedora-cisco-openh264’:


are you dl with tor?

it happens a lot, the dl just times out, keep trying or try new identity

with tor you have min 4 logical hops to dest compared to 1 logical hop if you just use ISP connection. each extra hop reduces reliablity

1 Like

That sounds reasonable for the mirror errors.

But, what might be causing the script errors such as those for the kernel, glibc, and NetworkManager packages? These don’t seem dependent on time.
Thanks for the prompt reply!