When you’re trying to enter and navigate websites, you may not always have a good experience. It all seems confusing sometimes, even daunting, with all the factors that have to be taken into account to make a web transaction successful. We’ve heard of such things as mashups, plug-ins, and mobile browsers. But for the average user, the basics are what count. It all comes down to availability and performance.

Ultimately, errors can be encountered related to the application, database, network, or the infrastructure. These areas require various technical disciplines, so finding a resolution to an outage sometimes requires lengthy conference calls to figure out whom to blame. Calls like this can be avoided, or at least shortened, with better monitoring. That’s why company websites should be thought of as end-to-end systems that focus more on what the user experiences and integrating web operations with a Business Service Management strategy (BSM).

Inevitably viewers will encounter issues while utilizing your website from time to time. Hard errors such as the commonly known 404 message, a non functioning database, network overload, and a loss of connection are fortunately easy to detect and clearly spelled out to a knowledgeable person. Website monitoring and pinging servers can catch these problems easily.

Soft errors are ones that cause the application to misbehave rather than break down entirely. If users have to log in twice, navigational paths are broken, page links and buttons don’t work, or necessary data cannot be submitted, a soft error has occurred. These errors are difficult to diagnose or detect, because users are confused and don’t know exactly what has gone wrong. What can help detect soft errors? Applications or transactions monitoring have the ability to detect such issues. How this works is a monitoring system will regularly execute a defined script that is programmed by the site owner. If there are any problems, the program will report these to the site owner.

Visitors to your site may have many different kinds of problems with your site’s performance. The most important measure of performance is the length of time required for the user to work with the application. Inconsistent performance can be even worse than slow performance in many cases. Users are likely to abandon your site and its applications if performance is unexpectedly poor, because they will believe that your site is broken when it slows down.

Latency or performance of websites can be affected by dozens of factors. These are six of the most common ones:

• 1. User-specific problems such as an obsolete computer or slow connection. There is nothing you can do.
• 2. Overall Internet congestion, which having the proper content delivery network and/or internet service providers can alleviate.
• 3. Server software that is also generating a report or doing something else that is resource-intensive, which intelligent software engineering can resolve
• 4. Insufficient infrastructure to handle the current load of traffic, which must be addressed with more hardware.
• 5. Application-specific problems based on your website or server.
• 6. Database performance that may be exacerbated by complicated data types and business logic operations, which database administrators should deal with.

The fifth generation of computing, the biggest thing since the web is cloud computing i.e., a means where data can be stored in web applications and accessed from anywhere and any browser. However, it is not without a backup plan. You need to safeguard all your data when you sense a storm coming with the help of these tools. Though advantageous with respect to storing and accessibility, it is disadvantageous as it is dependent on an external service to host, update, and maintain your much needed and beloved software. In doing so you place your data on computers that are not controlled either by you or at a single point of access resulting in failure of companies that can be shut down or bought, their accounts locked up, your servers and you become off-line. However, popular online services help you store your E-mail, pictures, files, address book, bookmarks, and journal entries in the cloud computing with easy ways to back up all your information from the cloud to your computer (learn more about free tools for backup here).

Availability (or Uptime) refers to the percentage of a specified period of time during which a computer system is capable of doing the things it is supposed to do. It came into use to describe the opposite of downtime, times when a system was not operational. For a networked system to be considered “available,” it has to have service, host and network working. It must have both connectivity (be physically attached) and functionality (actually working.) Typically, one reports availability as a single monthly figure. For example, 99% availability would mean that the system was not working for a little over 7 hours during the course of a month. Availability can be tested by sending a group of test packets and seeing if an answer is received.

Uptime — Time Lost in a year

• 98% – 7.3 days
• 99.0% – 3.7 days
• 99.9% – 8 hours
• 99.99% – 1 hour
• 99.999% – 5 minutes

Users of Microsoft Windows systems can type ‘systeminfo’ at the Command Prompt to display all system information, including the System Uptime. Users of UNIX and Linux systems can use the ‘uptime’ utility to get the uptime.

IT professionals are responsible for many tasks, but their main business objectives are to: align IT with the business and using IT to increase the efficiency of the business. In aligning IT with business, this requires the IT professional to make certain decisions and determine which actions to take that will best benefit the business. For operational efficiency, this means that while using the same amount of resources, service delivery is improved while lowering the costs at the same time.

In an effort of continuous improvement, IT must work to lower the cost of managing infrastructure. The traditional school of thought is to accomplish this through holistic, comprehensive approaches that are centralized in nature. This is the challenge that must be met in order to eliminate inefficiencies associated with silos and turf wars that result in fragmented IT management practices, and ultimately drive up the costs of system management.

To be effective, the IT system management infrastructure must: 1) Simplify and streamline system management through automation 2) Enable IT to perform proactive system management, heading off potential problems before they result in service degradation 3) Enable IT to prioritize system management activities based on business impact 5) Permit integration of IT processes across IT service management disciplines.

In order to meet agreed upon service levels (SLA’s) the system infrastructure must contain a strong monitoring component. This monitoring component, in order to be effective and efficient, must be done in a comprehensive manner from a centralized perspective. In addition to meeting agreed upon SLA’s, this will allow IT to minimize initial and resolution response times for any and all business critical services.

In the past, IT monitoring infrastructure was either agent-based or agentless, and IT companies were forced to choose between the two. Now, thanks to innovative new technology, the advantages of agent-based and agentless monitoring can be combined into a single, superior system. Using Monitis, IT organizations can choose the monitoring system that best fits their needs and budgets.

Monitis offers a fresh hybrid approach that provides the advantages of both agent-based and agentless monitoring. With the latest hybrid monitoring approach, IT can merge agent-based and agentless monitoring technologies to produce a single, integrated enterprise monitoring infrastructure. It provides the vastly granular monitoring presented by agent-based technology where required, combined with the cheaper cost of agent less technology, where in-depth monitoring is not needed. For example, the IT staff can apply agent-based monitoring on crucial ERP and Web servers, and put implement lightweight presence or agentless monitoring into action on less significant servers and department printers. This provides IT a complete and affordable view of the whole IT environment. As an outcome, the IT Staff will achieve operational visibility into all components of the IT environment, monitoring components that were not previously monitored.

External Monitoring

• New wizard for adding external monitors(bulk add possibility)

Internal Monitoring

• Delete Agent names from the list
• Create New Tab and put the agent’s all  monitors into it
• Agent Snapshot

Widgets and Tools

• Command-Line Tool Easymonitis
• VT widget – Automatically add a module to the user’s default tab when adding a test
• VT widget – Automatically add an external monitor when adding a VT test and show uptime in VT widget

General

• Schedule Maintenance
• Notify When Back Up
• Amazon Payment System
• Advanced User Preference Management Mechanism
• Possibility to remove monitor by right from the dashboard
• Possibility to go to module’s tab whenever it already exists

External Monitoring

• New Monitoring location from China
• Possibility of excluding US location

Internal Monitoring

• FreeBSD agent

Transaction Monitoring

• Reliability and false positives improvements

EC2 Monitoring

• EC2 advanced policy

Widgets

• Sign up and add monitor feature in iGoogle Perfometer widget
• Sign up and add monitor feature in iGoogle Snapshot widget
• Sign up and add monitor feature in iGoogle Test widget

General

• Easy Start Wizard introduction
• Continuous alerting possibility
• Monifox version 1.2(Firefox add-on)
• Sending notification alerts to Twitter
• Coupons introduction during signup

We recently offered our Monitis users a 10-question survey to prioritize features for our roadmap. Our first question was about the dashboard user interface. We listed several planned enhancements and asked the users to point out which are the most critical for them. Users identified several performance chart views as the most critical needs like a 24-hours view, 7-days view widgets, and views with flexible custom dates. We instantly put these requests in our next week plan and hope to deliver them to our users within the next 2 weeks. We also included other requests in our roadmap.

The next question was related to external monitoring. One of the highest requests was to add checks for broken website links. We are actually set to launch this service by the end of the next week. The early release will be a free service, and later it will be integrated with Monitis. It will daily crawl the complete user website and will find dead-end links. Also, people asked for more monitoring locations. We are going to add at least one more location in a week.

The next block was related to internal, agent-based server and network monitoring. The highest request was for advanced MySQL database monitoring. We already have it developed and will integrate it in the nearest future. It will definitely be live by the end of June. The next priority was identified as adding customized scripts for probes. Technically it is also ready, so we will try to move the launch earlier in our schedule. The automatic recovery scripts feature was a winner in the nice-to-have category. Next, was the users request to integrate SNMP monitoring with Monitis (which is currently a separate service at www.monitorsnmp.com). Again, we will have them live by the end of June.

More about the survey results in our next blog post.

External Monitoring

• “User Agent” of external checks can be defined from Options, which will help to identify and filter requests from external monitoring probes in the logs. In some cases it will help to set up the User Agent the way which results in the correct response of the particular website.
• Calendar is added to the external monitors so it is more convenient to browser older data
• Embeddable public widget is added which allows to put it on your website
• Public report option is added, which provides a link to show your report to your visitors
• Improvements related to Maintenance Scheduling
• Possibility of selecting Australian location from the notification rules
• Possibility to deselect US location when adding new monitor and in the notification rules

Internal Monitoring

• Weekly reports will be sent every week showing statistics related to internal monitors
• “Agent is not running” alert can be set up from the Options in order to receive warning notifications in case something is wrong with connection between the monitored server and Monitis Central Server.

General

• Launch of new portal
• New type of monitoring is introduced namely – Amazon EC2 instance monitoring

Monitis has reached 500 Twitter followers at http://twitter.com/monitis . We use Twitter to inform our community about upcoming events and releases. Our users may also submit feedbacks and messages via Twitter.