Skip to content

Ericsson/ericsson-hds-agent

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Ericsson HDS Agent

What is Ericsson HDS Agent?

Ericsson HDS (Hyperscale Datacenter Systems) Agent is a Linux based program designed to run on any Linux System. It collects an inventory of the host machine's hardware inventory and runtime system metrics. It is composed of many built-in collectors, formatters and a forwarder.

image

Supported Platforms:

  • LINUX/x86_64
  • LINUX/ARM64

Recommended System Tools

The agent uses various Linux commands to collect inventory data from the host machine. Please refer to the documentation in docs folder for more details.

Getting Started

Installation from Binary Distribution

  1. Download

    Please go to the release folder to download the binary code.

  2. Install

    Extract the downloaded file to a directory from where the user wants to run.

    tar -zxvf ericsson-hds-agent_x86_64.tar.gz
    
  3. Run

    After the files are extracted, the program can be executed as follows:

    cd ericsson-hds-agent/
    
    sudo ./ericsson-hds-agent -stdout
    

    A node.id file is automatically created and contains an unique identifier for this host system.

    The above command runs the inventory and metrics collectors once and sends the collected data to standard output.

    This behavior can be changed by passing different sets of command line arguments. Here are some examples.

    sudo ./ericsson-hds-agent -stdout -frequency=15
    

    The above command collects metrics at an interval of 15 seconds and sends the data to standard output.

    sudo ./ericsson-hds-agent  -frequency=15 -destination=tcp:<ip_addr>:9090
    

    (ip_addr = ip address of the storage server.)

    The above command collects metrics at an interval of 15 seconds and sends the data to a storage server running on port 9090. A sample storage server has been provided to ingest the data.

Command Line Arguments and Other details

Please refer to detailed document for a complete list of supported parameters and other details of the Ericsson HDS Agent.

Installation from Source Code

Prerequisite:

  • git
  • go 1.6

Follow these installation and setup instruction links git and go 1.6 if you neeed to install.

  1. How to Download

    User can download the Ericsson HDS Agent source code using go command

    mkdir -p $HOME/go/src
    
    export GOPATH=$HOME/go
    
    cd $GOPATH/src
    
    go get github.com/Ericsson/ericsson-hds-agent
    
  2. How to Build

    Execute the following set of commands to build HDS Agent binary:

    cd $GOPATH/src/github.com/Ericsson/ericsson-hds-agent/agent/apps/ericsson-hds-agent
    
    go get ./...
    
    go build
    
  3. How to Run

    After the build completes, follow this instruction to execute the Ericsson HDS Agent program:

    sudo ./ericsson-hds-agent -stdout
    

    After the agent is run, a node.id file is automatically created. This file contains an unique identifier for this agent. To run with other parameters refer to section Running the agent

  4. How to Validate Data Being Collected

    User can validate Ericsson HDS Agent output with system command(s)

    For example, cpu inventory data output from Ericsson HDS Agent(shown below) can be verified using Linux command cat /proc/cpuinfo.

     {
         "Category": "cpuinfo",
         "Details": [
     	{
     	    "Tag": "processor",
     	    "Value": "0"
     	},
     	{
     	    "Tag": "vendor_id",
     	    "Value": "GenuineIntel"
     	},
     	{
     	    "Tag": "cpu family",
     	    "Value": "6"
     	},
     	{
     	    "Tag": "model",
     	    "Value": "70"
     	},
     	{
     	    "Tag": "model name",
     	    "Value": "Intel(R) Core(TM) i7-4870HQ CPU @ 2.50GHz"
     	},
     	...
         ]
     }
    
  5. How to Use the Data
    In this example, the inventory and metrics data from the Ericsson HDS Agent can be viewed on a dashboard to analyze the inventory of the data center and how each machine is operating.

image image