Monitoring Service
This service monitors the health of your DCS servers.<p> It detects crashes or unwanted popups, full disks, too low server FPS or too high RAM usage. It also generates load information to be used with the /serverload
command.
Servers that are considered RUNNING, PAUSED or STOPPED will be monitored for any unusual popup (login or lua error) or crashes of the respective DCS.exe or DCS_server.exe process. A heartbeat will be sent to DCS every minute. The maximum number of heartbeats a server can miss can be configured in your instance configuration in nodes.yaml
Configuration
The (optional) configuration file for this service has to be placed into config\services\monitoring.yaml:
DEFAULT:
time_sync: true # sync the PC time with a time-server every 12 hrs, default: false
time_server: pool.ntp.org # and use this non-default time-server for it, default: Windows default
thresholds:
Drive: # You cannot disable the drive check. If you do not specify anything, these values will be taken as default.
warn: 10 # Warn, if your drive where DCS is installed (or your C: drive), gets below 10% (default: 10)
alert: 5 # Send an alert if your DCS drive (or your C: drive) gets below 5% (default: 5)
message: "Available space on drive {drive} has dropped below {pct}%!\nOnly {bytes_free} out of {bytes_total} free."
FPS: # Optional FPS-check
min: 30 # if FPS reaches a min of 30 (default: 30)
period: 5 # for at least 5 minutes (default: 5)
message: "Server {server} FPS ({fps}) has been below {min_fps} for more than {period} minutes."
mentioning: true # and mention the admins (default: true)
RAM: # Optional RAM-check
max: 32 # if RAM exceeds 32 GB (default: 32)
period: 5 # for at least 5 minutes (default: 5)
message: "Server {server} RAM usage is {ram} GB, exceeding the maximum of {max_ram} GB for more than {period} minutes."
mentioning: true # and mention the admins (default: true)
To configure the DCS server monitoring, you can change these values in your nodes.yaml and scheduler.yaml:
nodes.yaml
MyNode:
DCS.release_server:
max_hung_minutes: 5 # maximum heartbeats a server can miss (default: 3)
plugins/scheduler.yaml
DCS.release_server:
schedule:
00-24: YYYYYYY # the server (and its DCS_server.exe process) should run 24x7