As part of our testing procedure, we verify that the network connectivity to your server is open and performing. Before we test your server itself, we perform an advanced TCP/IP connectivity test that measures the response time of your network. We feel it is just as important to know how your network is performing as it is to know how your server is performing.
As part of our testing procedure, we verify that your DNS configuration and servers are working properly. Before we test your server, we analyze the DNS response at the perimeter of the network to verify that your DNS records are being distributed across the internet and cached on remote DNS servers effectively. This system of DNS distribution must be functioning properly for people to access your server using your domain name (ex. http://www.mysite.com).
As part of our testing procedure, we send out a Ping to your server to measure its responsiveness on the network. We then save this ping data and make it available to you in performance reports. These performance reports allow you to see trends and patterns of network performance over time.
As part of our testing procedure, we verify that the TCP-layer of your network is operational. We do this by making a TCP connection to an open port on your server and verifying that the TCP handshake is successful.
You can use our Test Profiles feature to test multiple server access scenarios. For example, you can test both SSL & non-SSL, multiple domain names for the same server, and multiple input variations. Your monitor will cycle through each of the scenarios in a round robin fashion (testing one scenario per test time). For best results, this feature should only be used to test server components that rely on the same backend server, application code and database. To test different servers, a new monitor should be created for each.
By default, all accounts are tested from multiple test locations in our core network. Monitoring from multiple geographic regions allows you to measure how your server is performing for users in those different regions.
AlertBot is designed to be as specific as possible when diagnosing problems with your server. Our system is able to identify many unique Error Signatures and report them to you by name. This helps you diagnose problems quickly and more efficiently when you receive a failure alert.
When we detect a problem with your server, we immediately take appropriate measures to validate and diagnose the error. The first thing we do is immediately retest your server from a different geographic location to validate that there is an actual problem with your server and not just a false alarm. We then perform an advanced TCP/IP test that helps identify problems with routers, network equipment, and NIC cards connected to your server. We continue our diagnostics by validating the response of the web applications running on your server. We then attempt to capture and uniquely identify the errors your web server is throwing. These are just a few of the many diagnostics we perform to troubleshoot your server problems.
One of the most important features of our system is its advanced false alert protection. No one likes to be alerted and have it turn out to be a false alarm (particularly in the middle of the night). AlertBot goes to great lengths to protect against false alerts. One of the key processes we use to protect against false alerts is that we verify failures from multiple independent test stations located in completely different geographic regions. We then compare the results from these locations and only classify the failure as valid if it's corroborated by each of the other test locations. This ensures you're not bothered with an alert unless it's truly a problem with your network.
Contact escalation lets you configure the time each contact should be alerted during the course of a web site or server failure. For example, you can have a technician alerted immediately. If the web site remains down after 10 more minutes, you can have two more technicians alerted. You can then have an upper level administrator contacted if the error still persists after 20 minutes and so on. These escalation times can be customized to any quantity of levels, times and contacts you see fit.
You can setup any number of cell phone or mobile devices for text message based alerts. Whenever your server fails, we will send you an alert that will show up on your phone as a standard text message. This text message will include information about what server is failing and the name of the error it's throwing. It will also include information like total downtime and the date/time of the last test. You can receive an unlimited number of SMS Alerts (via email gateway) from us for free. Please note, however, that your cell phone provider may charge you for text messaging.
You can setup any number of email addresses for alerts. Whenever your server fails, we will send out an email to each of these addresses alerting you of the failure. These email messages will include information about what server is failing and the name of the error it's throwing. They will also include information like total downtime and the data/time of the last test.
If email and SMS alerts are not enough, we also offer direct phone call alerts where our automated system will call you and speak the alert information to you out loud. If our system is not able to reach you on the first call, it will re-attempt the call multiple times.
We allow you to schedule recurring maintenance windows where your monitoring is temporarily disabled so you're not alerted of expected downtime. This is most commonly used when you have maintenance being performed on your server on a recurring basis and you expect your servers to be down during that time.
Our monitoring service not only verifies that your server is functioning without error, but it also takes performance measurements during its test procedure. Performance measurements include connection time, response time, download time, and transfer rate. This data is stored and made available for your analysis in our Reports section. These reports are extremely useful in identifying potential bottlenecks and performance problems.
During each test performed on your server, we collect and store various pieces of performance data. We then take this performance data and make it available to you via easy to read performance reports. These reports include ping response time charts, test step breakdown charts, TCP response time charts, transfer rate charts and more. You can run these reports over hours, days, weeks or months, and see the overall performance trends and patterns. These performance charts become invaluable in determining bottlenecks in your network and web capacity during peak hours and network backup windows.
Reports are delivered through an interactive charting interface that allows you to customize reports. You simply choose the data that you want to look at, and the type of chart you want to display it on, and you're on your way. You can choose from 20 different types of charts-- like Bar, Line, Area, Pie, Scatter, Radar, Surface and Contour-- to display your data. In addition, there are many more features that you can customize through the intuitive menus and toolbars. You can even export charts and data to formats like Binary, Metafile and Bitmap, and delimited Text files. Our reporting system puts the power in your hands, rather than limiting you to pre-defined reports.
When you add users and contacts to your account, we allow you to specify what time zone they are in. This information allows us to convert dates and times displayed in reports and alerts to your user's relevant time zone. This way you never have to worry about manually converting dates and times in our system to your local time.
We provide a test log, which lists the information about every test performed on your server. This test data includes the date/time of the test, the geographic location where it was performed, the status of each test step and the individual performance data taken for each test step. The test log can be filtered by monitor, test profile, geographic location, test errors, and more. The test log is a great tool for analyzing the raw test data.
We provide a detailed Failure Event Report for each confirmed failure that occurs with your server. The Failure Event Report allows you to see all the details regarding a failure in one place. The Failure Event Report includes the start and end date/times of the failure, a list of all the tests performed during the failure event, a list of the geographic locations where the failures were detected, and a list all the unique errors identified during the failure event. These Failure Event Reports are listed in the Failure Log. The Failure Log allows you to filter the list of Failure Reports by monitor and date/time so that you can quickly find the report you're looking for. Among other things, Failure Event Reports can be used during management meetings and event debriefings.
We provide a report which allows you to clearly see which errors are most common on your server. This report breaks down all errors that have occurred on your server into a doughnut chart with percentages assigned to each slice. This report is invaluable in determining the best place to spend development resources to reduce downtime. Starting with the most frequent offending error and working your way down to the least frequent offending error will produce the fastest reduction in downtime.
We provide a real-time monitoring status page which lists every monitor in your account and lets you know if it's up or down, when it was last tested, the performance results of that last test, and what date/time the next test is going to be performed. This real-time status page allows you to get a quick view of your entire network all on one screen.
As with most companies, there may be multiple people in your company that need to interact with our system. We make this easy by allowing you to add multiple users to your account. Each user will have a different username and password and each user can have different assigned permissions.
You can assign a unique set of permissions to each user you add to your account. You can give each user permissions to view/add/edit/delete monitors, reports, test kit tools, account information, and users. This becomes useful when you need to give certain employees limited access to the system. For example, you might want to give some employees the ability to view reports while other employees can add/edit monitors but not delete them.
Quality you can trust;
Customer service
you can rely on!
Don't trust your monitoring to just anyone.
1-800-360-8092
Email
Follow us on Twitter
Become a Facebook Fan
AU
BR
CA
CN
DE
UK