With this change in place, the grafana service will signal
readiness to serve by writing "READY=1" to the path specified
through the NOTIFY_SOCKET environment variable. If this
environment variable is not present or empty, no notification
will happen. This notification is the standard systemd
mechanism for indicating a service is ready to serve. For
Grafana this may be a couple of seconds from startup due to
database migrations. This change also adjusts the Grafana
systemd service definition to make use of this feature.
In systemd `After=` merely codes an ordering relation and does not start
other units.
Grafana needs to specify that it wants to start up _after_ either of the
supported DB services have started, _if they start_.
Enabling the database services, however, is out of scope for Grafana.
With present postinst script on Centos 7, restart after upgrade is not working with Warning: grafana-server.service changed on disk. Run 'systemctl daemon-reload' to reload units.
So i added systemctl daemon-reload.
* get rid of weird line breaks and use action
When using restarts/stops/starts you'd get weird output sometimes, Strange line breaks 'OK' status overlapping the next lines etc...
This fixes those moves OK to the right place and colorizes them correctly.
* added : for more uniformity
when doing a restart the output looked like:
Stopping Grafana Server ... [ OK ]
Starting Grafana Server: ... [ OK ]
The Stopping line did not have a colon. I added it just to make it look better.