Multisystem Monitoring

You control access using IP address and user name-password combinations: RADIUS, LDAP or even Active Directory Services authentication would add value. To make up for it, though, SiteScope lets you manage multiple SiteScope servers, so you can distribute monitoring duties across several servers yet manage them centrally.

I tested a beta version of SiteScope on my Pentium 4, 1.8-GHz laptop running Red Hat Linux 8.0 with 512 MB of RAM. Within moments I was monitoring my Web server, Cisco switches, 3Com 802.11b access point, OpenLDAP Server and several Web services. To configure the system, I set up groups for the applications, services and devices I wanted to monitor: DNS, Web, Web services and infrastructure. SiteScope lets you create subgroups to further organize your monitoring; I made two subgroups under infrastructure, one for switches and another for applications.

Adding services to the appropriate group based on function and test parameters was easy. Adding Web services to groups was more complex. I added a .Net Web service, an Apache SOAP (Simple Object Access Protocol) Web service and a The Mind Electric Glue Web service; SiteScope monitored all equally well. Interoperability between the SiteScope agent and the Web services was smooth, but the method of configuring parameters is clunky.

First I named the Web services to be monitored, then entered the URL of the WSDL (Web Services Definition Language) file. SiteScope retrieved the methods available for monitoring and let me choose from them. I selected the appropriate parameters (or arguments) for the method, such as ZipCode, Name or e-mail, and chose their types, such as integer or string. I then assigned values to each parameter or argument.

The types of parameters are listed in a text area in which you assign values to each parameter "For .Net." This can be confusing: You'll have to sort through the XML and find the tag representing each parameter before you can edit it. RPC/Encoded Web services are a bit easier to configure.When specifying monitoring for groups, you need to go through a plethora of options to configure the setting of acceptable thresholds. Then you must define test schedules and actions to take based on the results of each test. The basic monitoring report shows response time and status. More detailed information is available via reports set up in each monitoring report.

You can configure the monitors to perform tests at any second, minute, day or week interval. Success is determined in any number of ways, including content, status codes returned and response time.

Technology editor Lori MacVittie works in our Green Bay, Wis., labs. Write to her at [email protected].