Site24x7 Server Monitoring (Agent Based)

Monitor your servers with our light weight, easy-to-install agent. Stay ahead of outages and performance degradation issues with over 60 performance metrics, intuitive dashboards, RCA, IT Automation, and instant alerting mechanisms.

• Agent architecture - U.S. data center
• Agent security
• OS and features
• Plugin integrations
• Reports, dashboards, & IT Automation
• Troubleshooting tips

Agent Architecture - U.S. Data Center

Learn how our server monitoring agent collects and displays data via three (one primary and two secondary) data centers, set up at different locations to ensure monitoring services remain uninterrupted in the event of a data center failure.

• Primary Data Center (PDC) - plus.site24x7.com
• Disaster Recovery Data Center (DRDC) - Primary (plus2.site24x7.com) and Secondary (plus3.site24x7.com)  

In case of any error in the PDC, status information via heartbeat check is automatically sent to the Primary DRDC, thus avoiding false alerts due to intermittent network failure. If there occurs an error in the Primary DRDC, the Secondary DRDC will still receive the heartbeat check to ensure continued monitoring. Server shutdowns under normal circumstances is clearly communicated by the agent to our data centers.

To establish a connection with our data centers, the agent uses dedicated IP domains and ports. We recommend adding them to your allowlist for seamless monitoring.

Data collection

Once the agent is successfully installed in the user machine, its communication to the PDC is initiated using a one-way HTTPS connection. The agent collects and sends three types of data.

• Heartbeat check: Performed every minute. Sends alerts if no data is received for seven minutes by default.
• Server metrics: Collected based on user-set intervals (one minute to one day). Data is pushed to data centers every five minutes. In case of network failure, data is stored in ZIP files and sent as backlogs when connectivity is restored.
• Device messaging system: Collects real-time data for actions like starting or stopping a process along with agent upgrades.

Heartbeat Check:

This is a status information check done every minute by the agent to the PDC. The server's uptime is decided by how long it's been since Site24x7 received a heartbeat check from the agent running on your server. If the agent is unable to communicate with the PDC, it will try contacting the primary DRDC and then the secondary DRDC.

If the heartbeat check has not occurred for a continuous time interval of seven minutes, the Server Availability Watcher will alert the user about possible network communication failure or system crash. You can also customize this time interval in the Admin > Server Monitor > Settings page and set an appropriate value for Set Threshold for Heartbeat Check. Learn more.

Data Collection with a Polling Interval of One Minute:

The server monitoring agent will collect data every minute, consolidate it and communicate it to the Site24x7 data center every five minutes. Only in case of any threshold violation, the agent will immediately communicate to the data center and trigger an alert to the user. In addition, you can also set polling interval from a range of 1 min, 2 min, 3 min, 5 min, 10 mins, 15 mins, 30 mins, 1 hour, 3 hours, 6 hours, 12 hours, and one day.

During a network connection failure, the data will be stored in the server as Zip files. Once the connection is restored, the agent will send the backlog data to the Site24x7 data center. Thus, there will be no data loss even during a communication failure.

Device Messaging System (DMS):

This ensures real-time communication to the monitored servers. Communication to the DMS happens and the agent performs several actions triggered by the user like starting/stopping a process, discovering a service/process, upgrading the agent, root cause analysis, adding a port, URL, file, directory, and event log/syslog checks. 

Server Monitoring Architecture

Network Sanity Checks:

Before reporting an outage, Site24x7 checks the availability status of its own internal network by performing a Network Sanity Check. This is to make sure that it is the resource that's down and not its network. Check out the possible outage scenarios and steps to troubleshoot them.

Agent's average resource consumption

The Site24x7 Linux Server monitoring agent v19.2.0 underwent resource consumption testing on a basic Linux server with a 1-core Intel i7 processor, 4GB RAM, and 100GB disk space. The results were as follows:

• 0.6% CPU utilization
• 130MB of memory utilized (RSS)
• 100MB of disk space occupied
• Maximum 1500 Bytes/s network bandwidth consumed
• 2477 calls to Site24x7 data center

Similarly, Windows server monitoring agent v20.6.0 was tested on a basic Windows Server 2019 Datacenter edition with a 1-core Intel Xeon Platinum 2.6GHz processor and 4GB RAM. The results were as follows:

• 0.15% CPU utilization
• 150MB of RAM utilized (to monitor one Windows application and three plugins)
• 250MB of disk space occupied
• Maximum 1500 Bytes/s network bandwidth consumed
• 3750 calls to Site24x7 data center

These results were obtained with the agent running without additional plugins, log management (AppLogs) services, or applications in the specified environment. Changes to configuration or environment may impact resource consumption.

Logs generated by the agent

Agent operations are logged and stored in the following OS-specific locations:

• Linux: <Installation_Directory>/site24x7/monagent/logs
• Windows: <Installation_Directory>/Site24x7/WinAgent/Monitoring/Logs

Learn more about the description, maximum size, and the count of each log file here.

Upgrade the agent to the latest version

Upgrade the server monitoring agent either automatically through Site24x7's web-client in the settings page or manually by following our installation procedures. Learn more about our agent upgrade process here.

OS specific agent details and actions

Windows | Linux

Windows

Agent services

There are five services run by the agent:

• Site24x7 Agent Helper: Manages the Windows agent
• Site24x7 APP Monitoring Agent: Collects application data
• Site24x7 Applog Agent: Collects logs
• Site24x7 Plugin Agent: Integrates plugins
• Site24x7 Windows Agent: The primary Windows server monitoring agent

Supported OS versions

• Windows: 7 and above
• Windows Server: 2008 and above

Agent actions

Perform agent actions from the system tray. Open the system tray to find the Site24x7 icon. Right click to perform the following agent actions:

• View agent logs
• Perform connectivity check
• Access troubleshooting tips
• Start agent
• Stop agent
• Restart agent
• View the proxy configuration
• View agent details
• View agent status
• Bulk install the server monitoring agent

Linux

Agent processes

There are five services run by the agent:

• Agent Watchdog: Manages the Linux agent
• Applog Agent (optional): Collects logs when AppLogs is enabled
• Metrics Agent (optional): Runs when StatsD and Prometheus are enabled for monitoring
• Linux Agent: The primary Linux server monitoring agent

Supported Linux flavors

Agent actions

Perform agent actions from your terminal. Run the following commands for performing agent actions. Replace with the preferred agent action command parameter.

Agent installed as root

• /opt/site24x7/monagent/scripts/AgentManager.sh <option>

Agent installed as non-root

• /<installed_directory>/site24x7/monagent/scripts/AgentManager.sh <option>

MySQL monitoring specific actions

Agent Security

As the agent has to be downloaded and installed on every server that needs monitoring, it is important to understand about its security, its resource usage and the prerequisites for successful installation. Learn more.

OS and Features

Learn more about the platforms supported, the bulk deployment options available for installing the agent in 100's of servers in one go, and the various performance metrics provided for each of these platforms.

Plugin Integrations

Monitor your databases, applications, web servers, caches, message brokers and keep track of data that matters the most. Take a look at our 100+ out-of-the-box plugins. If you don't find the one you wish to monitor, write your own using Python and Shell script for Linux and VB, Powershell, Batch and DLL for Windows.

Reports, Dashboards, & IT Automation