# fleet

Jesse Mundis

05/10/2023, 5:59 PM
Best practice question. I have fleetDM deployed in a docker container on an EC2 instance. The docker launch sets a bunch of variables and flags and such. If I want to cleanly shutdown and restart “fleetDM” is
docker restart fleet
best practice, or is there some better way, say with ‘fleetctl’ to restart the server? I’m concerned that database connections and state are all shut down and brought back cleanly, with any pending cache flushed to disk. Thanks.

Kathy Satterlee

05/10/2023, 6:05 PM
@Jesse Mundis Since Fleet uses Redis to manage cache, you're good to go with
docker restart
Since it's very common to use autoscaling and multiple Fleet instances, things are set up so that the server can just be killed without causing issues down the road. There was a potential "gotcha" with scheduled tasks if the server went down while a
job was running, but that has been resolved as of v 4.29.0.

Jesse Mundis

05/10/2023, 9:12 PM
Thank you both