Server Monitoring

AggreGate Network Manager provides a consolidated, comprehensive solution for monitoring your servers regardless of their hardware and platform.

Monitoring is performed using SNMP, WMI, Telnet/SSH, and other standard technologies. The following operating systems are supported out-of-the-box:

Windows FreeBSD AIX
Linux/Unix Solaris HP-UX
Mac OS
Server monitoring via SNMP, WMI, Telnet/SSH, and other standard technologies.

Server Discovery

The discovery process finds all servers in your network and auto-detects their type. All networked applications and services are discovered by the port scanning even if a server doesn't respond SNMP and even ping requests.

Server Performance and Health Monitoring

AggreGate Network Manager provides the alerting, charting, reporting, long-term history storage, and deep analysis options for all critical server health metrics:

CPU load
Disk space usage
Traffic and bandwidth usage
Running processes and services
CPU/HDD/motherboard temperature
FAN speed and status
Any custom metrics available via SNMP

Built-in alerts are provided for the important and critical server events, such as Authorization Failure, Superuser Login, Kernel Alert/Emergency, etc.

All alerts can be followed by the corrective actions, such as shell script execution or Wake-on-LAN.

An integrated task scheduler helps set up periodic emailing of daily, weekly or monthly reports to the system administrator or company executives.

Agentless Monitoring

AggreGate Network Manager monitors the network without any agents, so there is no additional software or hardware required to install on the target servers.

Application/Service Monitoring

Applications and services "known" by AggreGate Network Manager are monitored via the dedicated "smart" meters. This assumes checking the application operability instead of its availability.

Most service-specific meters have a similar behavior:

Connection to the application/service via its native protocol
Passing the authentication/authorization procedure
Retrieving the application state and key metrics
Sending some pre-defined requests
Caching the application reply within the system core for further analysis

Windows Event Log and Syslog messages from the diverse applications are collected and stored in a database. These messages can also trigger the pre-built alerts, such as Mail Emergency, FTP Login, etc.

Generic TCP/UDP Service Monitoring

Operability of the unrecognized applications and services (both TCP and UDP) can be monitored by:

Establishing a connection with a service (for TCP-based services) and checking the service availability
Sending some raw input data
Reading the raw output and injecting it into the system core for analysis
Measuring the response time and storing its stats

Database Monitoring

AggreGate Network Manager monitors the database servers via JDBC/ODBC. Virtually all modern enterprise databases are supported, including:

Microsoft SQL Server

The system is capable of executing dynamically constructed arbitrary SQL queries and converting the result sets into the AggreGate format to make them available for analysis. Dynamically constructed insert/update/delete queries can be also executed on-demand.

File/Folder Monitoring

File and folder periodic monitoring operations include:

Checking the file/folder existence
Validating the file size, modification time and checksum
Verifying the number of files in a folder and their total size
Loading the file contents into the system and making it available for editing/analysis
Folder contents inspection

Custom alerts may be triggered if the above operations lead to some unexpected results.

Remote Script Execution

AggreGate Network Manager can execute the custom shell scripts and applications on remote machines over a secure SSH connection to allow the acquisition of non-standard data and execution of different actions (e.g., rebooting the servers).

The scripts may be executed:

Periodically, according to the custom schedule
Upon alerts

Physical and Virtual Infrastructure Monitoring

The system makes no difference between the physical and virtualized servers. Guest operating systems running under the VMWare ESX/ESXi servers provide the same metrics as their "real" counterparts.

Based on Linux, the VMWare servers can be also monitored like any other physical machines, providing the same set of performance indicators. However, the virtualized infrastructure monitoring and management involve a lot of dedicated tasks, e.g. starting, stopping or suspending the Guest VMs.