Operating NEO Services
NVIDIA® NEO® users may start, stop, or restart NEO services, or check their status at any time.
To start Mellanox NEO services, run:
/opt/neo/neoservice start
In order to stop NEO services, run:
/opt/neo/neoservice stop
In order to restart NEO services, run:
/opt/neo/neoservice restart
In order to check NEO services status, run:
/opt/neo/neoservice status
NEO uses Monit to monitor the status of NEO and dependent services (influxdb, telegraf, kapacitor). If one of these services is down, Monit detects it and restarts the service after a few seconds.
To see the exact monitoring configuration, please refer to /etc/monit.d/neo.monitrc.
NEO incorporates a monitoring mechanism that can be combined with MCare, a support program that offers 24/7 fabric management services to monitor network health. This mechanism traps network events and issues regular notifications to the Network Operations Center (NOC). Specialized NVIDIA personnel analyze the details of the reported events and take action according to the service level agreement (SLA).
MCare identifies, alerts and addresses hardware failures, non-optimal configuration, service degradation issues, performance issues and more.
To obtain an MCare license, please contact your NVIDIA Support.