EventSentry can monitor a variety of software applications through its log, event log and system health monitoring functionality. Its flexible ruleset and wide variety of notification types create an extremely powerful and flexible monitoring environment.
Errors can be detected, forwarded to software developers or IT staff, and automatic corrective action can be taken.
The features listed below are suitable for application monitoring, and can be combined as you which to create monitoring “packages”, which can then be assigned to one or more monitored hosts.
Monitoring web-based applications can be done in a variety of ways, starting with the most basic TCP port check, to ensure the application/web service etc. is running. For more thorough tests of web sites, the checkurl utility can be used in combination with the application scheduler (see below). CheckURL verifies that a web page is up and running, is returning a valid HTTP code, and can optionally look for mandatory or unwanted text on the page. Since the application scheduler keeps track of execution times, the responsiveness of web applications can be measured and alerted upon as well. For example, get an alert if a specific URL takes more than 5 seconds to load, or if it returns a HTTP error code.
Many applications log critical errors to the Windows event log. EventSentry can monitor any Windows event log, including custom event logs, and notify any number of recipients. See Event Log Monitoring for a list of all event log related functionality. For example, certain error messages can be sent to developers, whereas others can be forwarded to a sysadmin. Logs can also be stored for future analysis, and different actions can be taken depending on when the error occurs (see review from MSDN magazine).
Monitor any text log file, and be alerted in real-time when a text matching your ruleset is found in a monitored file. Log files can also be consolidated into the central EventSentry database, and delimited log files can be mapped inside EventSentry to support structured queries through the web reports. Log files can be specified using wildcards, so that monitoring files which are added dynamically to directories is as easy as possible. See Log File Monitoring for more information.
EventSentry’s applications scheduler is a powerful way to integrate, organize and execute custom monitoring scripts. Any script (e.g. VBScript, Powershell, Perl, Python, etc.) executed by the application scheduler will log its duration and output automatically to the application event log, including the return code (%ERRORLEVEL%). From there, custom event log filter rules can be applied, and alerts be dispatched. Scripts can also be managed inside EventSentry, so that they do not need to be manually maintained on the monitored servers. See example of scripts.
If the monitored application runs as a service, then EventSentry’s service monitoring can alert you when the status of the service changes. A history of service status changes can also be collected in the central reporting database.
If the application is just a process, then EventSentry’s process monitoring feature can also issue an alert when not enough instances of the process are active (you can require one or more instances of the process to be active).
Whether or not the monitored application exposes performance counters, the performance monitoring feature will ensure that monitored applications will not consume more resources than intended. Windows lets you monitor a variety of aspects of a process, including CPU usage, memory usage, handle count and more. Performance information can be consolidate in the central database, so that historic data can easily be analyzed through a variety of ways, e.g. with charts.
EventSentry can monitor any performance counter available on the system, including custom performance counters exposed by the monitored application.