Systemd service: Restart unit not found unless stopped first

Question

I have a systemd defined service, that cannot be restarted using systemctl unless I first do a stop. Systemctl claims that the unit does not exists. This happens everytime I reboot until I execute systemctl stop. How can I ensure that the service can be correctly restarted without first having to call systemctl stop?

Example:

# systemctl restart merlind.service
Failed to restart merlind.service: Unit not found.
# systemctl stop merlind.service
# systemctl restart merlind.service

Systemd service definition:

[Unit]
Description=Merlin
After=network.target

[Service]
ExecStart=/usr/bin/merlind --config /path/to/conf --debug
ExecStop=/usr/bin/merlind --config /path/to/conf --kill
Restart=always

[Install]
WantedBy=multi-user.target

The .service file is located at /usr/lib/systemd/system/ and symlinked at: /etc/systemd/system/multi-user.target.wants/

CentOS Linux release 7.5.1804 (Core)

Systemd version: 219

Answer 1

In this case, I had a another service, lets call it ServiceA, which required Merlin described here, as well as some other services. However, one of the services which ServiceA required was missing from the system. Fixing this dependency, also fixed the problem with the Merlin service, which could then be restarted immediately at reboot, without first doing a stop.