On Mon, Feb 05, 2024 at 04:02:55PM +0100, Jiri Denemark wrote:
On Mon, Feb 05, 2024 at 14:42:09 +0000, Daniel P. Berrangé wrote:
On Mon, Feb 05, 2024 at 03:14:08PM +0100, Jiri Denemark wrote:
When checking for machined we do not really care whether systemd itself is running, we just need machined to be either running or socket activated by systemd. That is, exactly the same we do for logind.
That's not right. We very much *do* care whether systemd is running.
If systemd-machined is installed on the host but the OS is booted into sysvinit, then DBus will report that machined can be activated, but if it is activated then it certainly won't actually work.
I guess I didn't explain myself correctly, but this just changes our machind service check to do the same we do for logind. That is, machine1 service must be enabled (listed by ListActivatableNames). If it is, it must already be running (reported by ListNames) or systemd must be running.
Are you saying we should properly handle situations when a service we want to talk to is already running, but cannot work because systemd is not running? I would expect the service never be running in such case as it would just fail to start without systemd.
Hmm, I guess that's safe enough actually. With regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|