Monitoring

Appliance resources can be monitored at a system-wide level. Exhaustion of any of the resources can have a negative impact on the speed of the transcription.

The following resources that can be monitored:

Resource ID (rID)	Description
cpu	Provides the CPU usage across all the vCPU assigned
mem	Provides the total RAM usage of the appliance

Here is an example GET request for the mem (RAM) resource:

curl -L -X GET "http://${APPLIANCE_HOST}:8080/v1/management/resource/mem" \
    -H 'Accept: application/json' \
    | jq

Here is an example response:

{
    "rId": "mem",
    "percentage": 13.4
}

For advanced monitoring, a utility called Glances is available that runs on TCP port 3000. It allows real-time resource stats to be monitored on the Real-time Virtual Appliance. The easiest way to access this is via a web browser using the link http://${APPLIANCE_HOST}:3000/ in the address bar.

Glances view

It is also possible to access the Glances API using XML-RPC or HTTP REST (for JSON output), for example:

curl -L -X GET "http://${APPLIANCE_HOST}:3000/api/3/mem/percent" \
    -H 'Accept: application/json' \
    | jq

For more information on the HTTP REST interface, consult the Glances documentation.