Nowadays enterprises run multiple servers to provide various services. Performance of each of these servers is usually highly critical. Even if only one of these servers fails, it can affect the delivery of several business essential services.
NetCrunch provides multi-vendor support to monitor servers and critical applications continuously, along with their services and processes. It uses many different technologies and appropriate protocols for each monitoring target such as virtualization hosts, operating systems, and applications.
The program stores all the data for historical performance tracking and troubleshooting, thus eliminating the need for multiple server monitoring tools.
Usually, more than 50% of the issues are reported by the end-users. The server monitoring solution is the only way to detect early symptoms and pass relevant notifications to the IT team.
NetCrunch allows setting intelligent thresholds on system and application performance metrics. They can be defined as simple thresholds or by the deviation from baseline, change rate, etc. NetCrunch offers 8 types of thresholds. See more in the documentation
NetCrunch monitors various operating system parameters such as memory, processor and disk operations.
Read more in the documentation
@@img:detailed-charts.gif Server Performance Status
NetCrunch uses Monitoring Packs (sets of rules for monitoring) to manage the monitoring settings - many of them are automatically assigned for the given OS.
@@img:os-packs.png OS Monitoring Packs
NetCrunch offers a quick insight into the operating system with System Views. The views depend on the operating system.
For example, for Linux family systems (MacOS, BSD, Solaris) NetCrunch presents list of logged users, opened files and network connections.
On Windows machines, you can access a list of services, processes, sessions and pending updates.
NetCrunch uses a remote connection to the Windows system to monitor its performance (no SNMP or agent required) by retrieving performance counters available in Windows.
Windows monitoring relies on a watching of Windows services, Windows Event Log, and perfmon counters. Additionally, you can use WMI sensors for in-depth access to system data.
NetCrunch offers a quick insight into Windows machines with System Views such as:
@@img:system-views.png Windows Services
NetCrunch can monitor windows service state based on the service state and startup type. For example, you can be alerted only when the given service is not running and is not disabled.
NetCrunch allows you to define simple alert filters to convert event log events into NetCrunch alerts. These filters are automatically converted into appropriate WQL queries.
NetCrunch provides 13 WMI sensors out of the box:
@@img:linux-processes.png Linux Process List
NetCrunch supports agentless monitoring of Unix-based systems using the SSH protocol. This allows you to avoid the problematic usage of SNMP on those systems. Using SNMP is always an option.
NetCrunch can monitor various Linux distributions based on kernel 2.4 or newer. It has been tested with: Cent OS, RedHat, Fedora, Novell OES, Ubuntu Desktop, and Server. It also supports various BSD edition such as FreeBSD, OpenBSD, and NetBSD.
NetCrunch monitors all vital system parameters and contains ready to use Monitoring Packs for each system.
See Monitoring Packs
NetCrunch offers advanced event management, including conditional alerts and correlation. The most important view for the administrator is Pending Alert View which shows only alerts that are not resolved yet. In this way, he can focus on current issues, without a hassle of browsing a history log.
@@img:event-log-pending.png Event Log Pending Alerts
Read more about Alerting, Escalation and Event Log Management
NetCrunch has also the ability to parse any kind of text logs and includes parsing expressions for custom log processing.
NetCrunch supports ESXi version 5.5, 6 and 6.5. It connects directly to the ESXi servers, so it does not need vSphere to be installed. This allows skipping single point of failure that can be a connection to vSphere.
NetCrunch comes with pre-configured Automatic Monitoring Packs to monitor ESXi as soon as OS monitoring is set to ESXi.
@@img:esx-hardware-status.png ESX Hardware Status
SQL sensors allow measuring connectivity, query execution, and can process result data as metrics or statuses.
Allows executing query returning multiple rows. Each numeric column can be used as a source for performance metrics.
This type of query should return a single row. It can be used for checking database authentication and connectivity. Returned data row can be treated as a status object.
Supported databases: SQL Server, Oracle, PostgreSQL, MySQL, MariaDB, ODBC source
NetCrunch offers several sensors for web/cloud monitoring.
It can send simple HTTP requests to measure single request-response time or load and render a whole web page as a web browser does. Finally, it supports REST requests and data sensors allowing processing of external data.
The sensor renders the page like a web browser, it loads all resources and runs scripts. It's intended for monitoring modern web pages or applications. It supports standard login and custom login forms.
The sensor loads data from a file using one of the predefined file formats. It supports native NetCrunch XML, JSON, and CSV and also allows to process data through Data Parsers which support consuming any data.
This sensor sends a single request and can alert on response code or by checking response data. It can send GET, HEAD and POST requests.
Similar to Basic HTTP sensor, this sensor can also send a single request and can alert on response code or by checking response data. Additionally, It can send any type HTTP request including a custom request and allows easy adding of URL query parameters. It also allows setting custom headers and cookies for the request.
@@img:web-page-sensor.png WebPage Sensor Settings
They enable efficient management of the monitoring settings. They provide settings for monitors based on object/counter data model such as SNMP or Windows PerfMon. Built-in agents provide such data.
You can use them to create monitoring policies by setting node filters. They can also be assigned manually to the particular node (or multiple nodes using multiselection). Currently, the program includes more than 180 ready to use Monitoring Packs for monitoring devices, applications, and operating systems.
They are similar to Monitoring Packs as they consist of default alerts and data collectors but are much more complex.
Sensors are dynamic monitors performing various actions to measure or retrieve remote data.
They offer more monitoring power in exchange for a more complex setup process.
The sensor usually requires a definition of the data source or actions to be taken (credentials, queries, paths, and parameters). For example, SQL Query sensor needs a database connection profile and an SQL query.
Read more about what can be monitored at What can I monitor
You can find an actual list of sensors here: Sensors
NetCrunch receives syslog and web message by default, even from nodes that are not in the Atlas. All you need to configure is the extenal source to send a message to NetCrunch.
Turning received messages from External Events windows into alerts is easy, the program will even create a new node if it is not defined in the Atlas yet. You can also define more complex alerts by filtering message content.
NetCrunch receives webhook message via HTTP POST request. You can quickly send notifications to NetCrunch using Amazon SNS service or you can use
curl from a command line.
Read more in External Event Sources
With NetCrunch you can start monitoring in minutes. Discovery and policies make initial setup a breeze. It can be a bit too fast for some. We quite often hear users asking: "That's it?". Yes indeed. We value your time.
NetCrunch is a unified product, built from the beginning as the general, system-agnostic and integrated solution.
NetCruch 10 introduced a new fast and scalable sensor engine. NetCrunch already has 8500+ MIBs in the database, and the list of Monitoring Packs and sensors is growing with every version.
Here at AdRem Software, we believe in "do it right or not at all" so we deliver professional solutions instead of workarounds.
In our tests, we try to break NetCrunch every day by overloading it, pushing the performance limits a little further with every new version (similar to Formula 1).