Can’t abort the process while upgrading

[fbruetting]

Host system details

● ostree://fedora-workstation:fedora/31/x86_64/silverblue
                   Version: 31.20191007.n.0 (2019-10-07T08:23:03Z)
                BaseCommit: b2cc97efbe5399b740d53f7f5b8c50cfa6097f867e965545ad2147fd7ec85a33
              GPGSignature: Valid signature by 7D22D5867F2A4236474BF7B850CB390B3C3359C4
       RemovedBasePackages: gnome-shell-extension-window-list gnome-classic-session gnome-shell-extension-apps-menu gnome-shell-extension-horizontal-workspaces gnome-shell-extension-launch-new-instance gnome-shell-extension-places-menu 3.34.0-2.fc31, gnome-terminal-nautilus gnome-shell-extension-background-logo gnome-terminal 3.34.0-1.fc31
           LayeredPackages: GraphicsMagick android-tools ansible arm-image-installer baobab bat
                            borgbackup cheat chromium cockpit cockpit-machines cockpit-podman cryfs
                            dutree easytag exa fd-find fdupes fedora-workstation-repositories ffmpeg
                            file-roller file-roller-nautilus fuse-exfat gcc gnome-boxes
                            gnome-encfs-manager gnome-shell-extension-auto-move-windows
                            gnome-shell-extension-dash-to-dock gnome-shell-extension-gsconnect
                            gnome-shell-extension-media-player-indicator
                            gnome-shell-extension-suspend-button gnome-shell-extension-topicons-plus
                            gnome-tweaks gnome-usage gpick htop iftop julia
                            linux-libertine-biolinum-fonts linux-libertine-fonts mosh open-sans-fonts
                            powerline-go pv python3-matplotlib ripgrep setroubleshoot tilix-nautilus tmux
                            tokei xenlism-wildfire xenlism-wildfire-midnight youtube-dl zsh
             LocalPackages: rpmfusion-nonfree-release-31-0.3.noarch rpmfusion-free-release-31-0.3.noarch

Expected vs actual behavior

I’d like rpm-ostree respecting sent signals in a relatively short time, instead nothing happened when a few minutes ago, it was upgrading for a pretty long time (~10 minutes), obviously stuck and I sent SIGINT via Ctrl+C. Normally it works, but it seems that there are situations where it doesn’t.

[cgwalters]

It should always be safe to systemctl restart rpm-ostreed; potentially we should switch Ctrl-C to doing that if it's pressed a few times.

(Eventually we want to run txns in a separate process from the DBus daemon, which would be a more useful intermediate state)

[fbruetting]

I don’t know what txns is, but respecting common standards would safe users frustration and time. 😉️ Having to use systemctl rerstart rpm-ostreed doesn’t scale. 😜️

[jlebon]

At what part was it stuck? This is likely a dupe of #897.