Monitoring

This section contains a link to the infrastructure node monitoring application.

Monitoring is performed through the interaction of the following tools:

• Telegraf is an open-source agent used for collecting metrics and data from the system

• InfluxDB is a freely distributed time series database (TSDB). All data collected by Telegraf is sent to InfluxDB

• Grafana is a tool for graphical representation of metrics, compatible with the InfluxDB database. It is used for creating indicator panels (dashboards) that display specific metrics over a defined period of time

When first accessing the «Statistics» section, you will be redirected to the Grafana login page.

Please consult your Administrator to obtain the login and password for accessing the Grafana dashboard.

To view the collected metrics, follow these steps:

• Click on the «Dashboards» button to open the menu

• Select «Browse» to access the list of panels

• Click on the desired panel's name

By default, three sets of dashboards are provided.

Infrastructure Monitoring

This set is designed for monitoring configurable infrastructure units such as areas, regions, buildings, groups, etc.

This set includes the following dashboards:

• Server: region cluster, monitoring of Mediaserver service parameters such as version, CPU load, memory, socket statistics

• Common indicators, traffic statistics on media servers (overall and per each media server)

• Service Dashboard Server, statistics on cameras, streams, users, traffic, and viewers

• 3 days storage usage forecast, a three-day forecast of disk space utilization

• Disks health, monitoring of disk errors (HDD and SSD)

• System dashboard: region cluster, monitoring of server metrics (memory, processes, network, disks, etc.)

Alerts

This set contains dashboards for monitoring and alert rules configured for deviations from acceptable values.

• Close Wait alerts, monitoring of the close_wait parameter for all media servers and alerts in case of exceeding the threshold value

• Disk Usage alerting, monitoring of disk space usage and alerts in case of exceeding the threshold value

• Service state & input traffic, monitoring of incoming traffic to the server and alerts in case of decreasing below a certain value

Monitoring

This set contains dashboards for monitoring various performance indicators and service status in the Production installation.

• DHCP, monitoring of the DHCP service status and its operational statistics

• NS_APP Monitoring, monitoring of CPU and memory usage by resource-intensive processes

• Percona cluster, monitoring of the Percona cluster state (a three-node DB cluster) and alerts for individual node unavailability

• Supervisor processes monitoring, monitoring of processes launched through Supervisor on two App servers

• System dashboard, monitoring of server metrics (memory, processes, network, disks, etc.)