Live Monitoring with Grafana

🕓 10 minutes

What you’ll learn#

How to use the live monitoring of your application or environment to:

  • see the memory and cpu allocations,
  • investigate the network statistics,
  • create high-performance, fast and scalable applications.

Prerequisites#

Overview#

In CodeNow you are able to live monitor your environment and/or your application.

  1. If you want to see the live monitoring of the environment, go to "Environments" and choose the "My environments" option.
    • There, choose the environment, click on the "See the live monitoring" button and you will be redirected to the Grafana UI page.
go_to_my_env

  1. If you want to see the live monitoring of your application, simply go to "Applications" and choose the "My Applications" option.
    • There, choose the application, see the "Deployed Applications" section and click on the "See the live monitoring" button and you will be redirected to the Grafana UI page.
info

IMPORTANT NOTE: To be able to use live monitoring of the application, your environment should be named "prod".



live_monitoring

Description#

grafana_preview

  1. CPU Usage.
  • It is important to monitor the CPU usage, because you can control the flow of your application/environment.

  • If the CPU is overused, it can throttle processes and affect performance.

  • The graph shows you the CPU usage in real-time.

    • If CPU usage spikes upward, the user interface of that server will eventually slow down, and multiple processes will crash along with the application running on that server.
    • High CPU usage can also cause high memory utilization issues that can cause a server to go down.
cpu_usage

  1. CPU Quota.
  • This table represents the general information on CPU usage in a particular pod.
cpu_quota

  1. Memory Usage.
  • Monitoring memory usage is essential in guaranteeing maximum performance.

  • High rates of memory utilization results in decreased performance for the related processes.

  • The graph shows you the CPU usage in real-time.

    • A steady increase in memory utilization over time may indicate a memory leak. A memory leak is when memory is allocated by processes as they start, but is not released when they end.
    • Memory leaks degrade application performance over time. Typically, it may become unresponsive when memory is no longer available.
memory_usage

  1. Memory Quota.
  • This table represents the general information on memory usage in a particular pod.
memory_quota

  1. Current Network Usage.
  • Here you can see some of the main information on network usage, gathered in one table.
    • These details are shown for every pod of your environment.
    • You need to monitor these values, so your application performance won't be affected by any unexpected processes.
    • And if the productivity of your application was slowed down by any of the unwanted events, you are able to quickly analyze and fix these issues.
  • The detailed live monitoring on a single value is presented below.
network_usage

  1. Receive Bandwidth.
  • Receive bandwidth indicates how much of the receiving throughput capability is consumed by the node.
    • If your node consumes too much received bandwidth, it might indicate the "bottleneck problem".
      • That means your application requests will take longer than was expected.
      • It might end up with failed requests.
receive bandwidth

  1. Transmit Bandwidth.
  • Transmit bandwidth indicates how much of the transmitting throughput capability is consumed by the node.
    • If your node consumes too much transmit bandwidth, it also might indicate the "bottleneck problem".
    • Your application might fail to send the information using the network.
transmit bandwidth

  1. Rate of Received Packets.
  • This section shows the number of successfully received packets in a real-time graph.
received packets

  1. Rate of Transmitted Packets.
  • This section shows the number of successfully sent packets in a real-time graph.
transmitted packets

  1. Rate of Received Packets Dropped.
  • This section shows the number of lost packets that were supposed to be received in a real-time graph.
received dropped

  1. Rate of Transmitted Packets Dropped.
  • This section shows the number of lost packets that were supposed to be sent in a real-time graph.
transmitted dropped

What’s next?#

See our other manuals: