Monitis has added yet another video tutorial, this time showing how users can monitor their cloud instances on Amazon AWS, Rackspace, and GoGrid in just a few minutes. To view this and other tutorials or request a live demo, check our Demo page, and subscribe to our Youtube channel.

Previous videos showed how to set up internal monitoring on a physical server using the Monitis Smart Agent. Once installed, the agent collects and sends performance data to Monitis. The agent works on cloud instances too, but you’ll have to install an agent on each new instance. Cloud monitoring automates this process, so you can start monitoring your entire cloud with the click of a button.

The video shows how to configure an account with Amazon EC2. Go to Add Monitor>Cloud Monitor and select Amazon EC2. Enter an Account Name of your choosing, your Amazon AWS Account Number, Account Key, and Secret Key, all provided by Amazon when signing up. Next select the Amazon region you’ve signed up for. You’ll be prompted to upload your Amazon EC2 private key. Monitis will now authenticate you into your Amazon account. You can set monitoring and notification rules to specify what gets monitored on your servers and when alerts should be sent. Monitis can monitor your cloud instances both internally via agent for CPU, Processes, etc., and externally via SSH, HTTP, or Ping. Your Amazon instances will appear on your dashboard along with their performance statistics.

We received some great feedback about video tutorials for External, Internal, Transaction, and Full Page Monitoring, so we’ve decided to make more tutorials. This week we’ve added one for network monitoring.

For starters, you can easily Ping an IP address behind your firewall from a server on which you’ve installed our agent. Simply go to Add Monitor>Internal, check Ping, and enter the IP Address. You can set thresholds for # of Lost Packets and Packet size and set alerts to be notified by email, SMS, Phone call, or IM if these thresholds are reached.

SNMP, short for Simple Network Monitoring Protocol, is the most common protocol for checking network-attached devices, such as routers and switches, for conditions that warrant administrative attention. Once you have the Monitis Agent installed on one of your servers, you can configure SNMP on our web-based dashboard by going to Add Monitor>Internal, checking SNMP, entering the Host IP and the Object Identifier. Every network device comes with Object Identifiers that let you monitor certain characteristics of the device. You can also set up an SNMP Trap, which, instead of polling the device periodically, sends an urgent message to Monitis when a specific problem occurs. Watch as we set one of each of these monitors up in the video, and feel free to try yourself by signup up for our 15 day trial.

The third of our new video tutorials shows how to set up internal monitoring. With other products this is usually a time-consuming process, sometimes taking weeks. Servers usually need a VPN connection to your monitoring server. This can be a pain to set up, especially with remote servers, and particularly with open source systems like Nagios and Zenoss. You also need a database to store historical data and a reporting module. Monitis cuts setup time to 5 minutes by utilizing every shortcut enabled by cloud computing. Instead of reporting to your server, the Monitis internal agent sends encrypted data to the Monitis Cloud via https where it is stored on world class servers. Let’s walk through the steps outlined in the video to set up internal monitoring for 10 servers. We’ll time it. First we log into Monitis, click Add Monitor to load the internal monitoring wizard, and download the proper agent for our operating system. (30 seconds so far)

We are running Linux, so we can automatically install the agent on all ten machines. It takes about 3 minutes. If you’re running Windows, it takes 1 minute to install and activate (with your login) on each server. Here’s a Windows agent that’s been activated.

Firewalls don’t need to be touched since the agent only uses port 443. The same setup.zip file can be used to install the agent on multiple servers, so you only need to download once. Also, you’ll never need to upgrade agents because upgrades are done automatically without manual intervention. Next we go back to the internal monitoring wizard. A list of names assigned to your agents will load. You can assign a tag-name to group internal monitors together. This comes in handy when generating reports. We can multi-select all ten agents and select the metrics to monitor (CPU utilization, memory, drive, http, ping), and click Add. (45 seconds) Dozens of movable, re-sizable graphs appear on the dashboard. Soon they are populated with data.

It took just 4 minutes and 15 seconds to set this up. The time-saving features can be narrowed down to three main points.
1. Agents report to Monitis via https, no need to touch your firewall. Just install and you’re ready to go.
2. Upgrades are done automatically without your intervention. No more concern about patches or versions.
3. Internal Monitoring Wizard lets you configure through any browser, and bulk-configuration of agents is quick and easy.

This video focused on server monitoring. We have another video coming that focuses on network monitoring with SNMP, so stay tuned.

Starting this week Monitis is rolling out new video tutorials that will guide users through setting up each type of monitor. The videos are located under the Resources tab of our homepage.  One of the videos is for Full Page Load monitoring, which has been a huge hit since we added it to our services a few weeks ago. It allows users to analyze how objects in their webpage are loading in real browsers in different parts of the world and identify bottlenecks. This is an extremely important measure for understanding user experience, as studies have shown that even a slight lag in load time can cause a noticeable drop in viewership and sales. As the video highlights, Amazon.com discovered that a 100 ms increase in load time resulted in a 1% drop in sales.  One might consider 100ms an unnoticeable duration, but 1% of Amazon’s annual sales is a whopping $300 million! Load time is also one of the parameters used by Google in determining a website’s PageRank.

Adding Full Page Load monitors is very easy.  Simply go to Add Monitor>Full Page Load, enter the url, a name and tagname for your monitor, a timeout threshold, frequency in minutes, and the locations from which to monitor. Click add and it will be added to the dashboard instantly. You can edit settings and add notification rules by click the pencil icon at the top. Clicking a datapoint reveals how the objects loaded during a particular test.  Objects include CSS scripts, Javascript, individual images, RSS, redirects, Frames and iFrames. The load times of these components are shown in a graphical view sometimes called a waterfall, and also in a sortable table. For each object, Monitis lists the following quantities: HTTP Response Code, Total Download time, DNS, Connection, 1st byte, Content Download, and number of bytes.  Sorting these quanities allows the user to quickly identify bottlenecks in the webpage. For example, sorting the DNS column brings the objects with the longest DNS time to the top.

One of our users, a large online news service, found that the website was loading fast save for a couple banner ads that were taking longer than 30 seconds to load. A customer in the education space found that pictures were taking too long to load and needed to be compressed or switched to .png files.

So, below is the video for your viewing pleasure. And if you’d like pricing details check out our Plan Builder. Full Page Load monitoring starts at just $5/month for the same plan that some of our competitors charge $50/month for!

Setting up internal server and network monitoring can be a time-consuming process, sometimes taking weeks.  Servers usually need a VPN connection to your monitoring server.  This can be a pain to set up, especially with remote servers, particularly with open source systems like Nagios and Zenoss.  You also need a database to store historical data and a reporting module.  Monitis cuts setup time to 5 minutes by utilizing every shortcut enabled by cloud computing.  Instead of reporting to your server, the Monitis internal agent sends encrypted data to the Monitis Cloud via https where it is stored on world class servers.   Let’s do a timed setup of internal monitoring for 10 servers.First we log into Monitis, click Add Monitor to load the internal monitoring wizard, and download the proper agent for our operating system. (30 seconds)

We are running Linux, so we can automatically install the agent on all ten machines.  It takes about 3 minutes.  If you’re running Windows, it takes 1 minute to install and activate (with your login) on each server.  Here’s a Windows agent that’s been activated.

Firewalls don’t need to be touched since the agent only uses port 443.  The same setup.zip file can be used to install the agent on multiple servers, so you only need to download once.  Also, you’ll never need to upgrade agents because upgrades are done automatically without manual intervention.Next we go back to the internal monitoring wizard.  A list of names assigned to your agents will load.  You can assign a tag-name to group internal monitors together.  This comes in handy when generating reports.  We can multi-select all ten agents and select the metrics to monitor (CPU utilization, memory, drive, http, ping), and click Add.  (45 seconds) Dozens of movable, re-sizable graphs appear on the dashboard.  Soon they are populated with data.

It took just 4 minutes and 15 seconds to set this up.  The time-saving features can be narrowed down to three main points.1.  Agents report to Monitis via https, no need to touch your firewall.  Just install and you’re ready to go.2.  Upgrades are done automatically without your intervention.  No more concern about patches or versions.3.  Internal Monitoring Wizard lets you configure through any browser, and bulk-configuration of agents is quick and easy.

But don’t take my word for it, sign up for a 2 week trial and see how much time you’ll save.

Today, more companies than ever are offering complex business services over the Internet via cloud computing. For business-critical applications, every minute of downtime means lost sales, customers and opportunities. It is a challenge to keep websites up and running round the clock.

Web monitoring services can help businesses make sure that their sites are functioning. Yet, simply checking if a website is working and online doesn’t mean that users are having a satisfactory experience. It’s like driving a 20-year old car. Is it working? Well, yes. But are you enjoying the bumpy, rattly ride? That’s another story.

Due to many reasons like peak traffic at rush hours, publicity campaigns that drive people to sites, a bug in software, database issues or the failure of third-party services, your online applications can become slow or non functional, disappoint visitors and prod them to leave and visit rival sites.

The answer is transaction monitoring services, which have two goals:

1. ensuring fast application performance and good user experience for visitors from anywhere around the globe
2. guaranteeing complete business functionality of applications 24 hours, seven days per week

Package Delivery, Transportation and Logistics Case Study

Lets consider how a Monitis Inc., customer, the world’s largest package delivery company and a leading global provider of specialized transportation and logistics, uses transaction monitoring to ensure the continuous availability of billing and reporting modules on the company’s website. They’ve used Monitis to monitor the following activities on their site every 5 minute, 24 hours per day:

1. open the website
2. logging in to the system, using a “test” user account
2. go to the payments’ page and search for available invoices
3. find the latest invoice and go to the details’ page and check certain content
4. go and check the history of all invoices
5. go to the report generation page
6. generate report of payments and check the content
7. download the report
8. sign out from the system

Monitis Transaction Monitoring Summary Report

The above Monitis report summarizes the end-user customer experience for the company, reporting peak, mid-point and low activity periods. To investigate low-activity periods, the company studied a Monitis detailed transaction report, shown below, and discovered that its Sign In/Sign Out and Invoice History pages were slow. As a result of studying this information, the company was able to pinpoint operational issues, optimize its database and, in the end, improve the user experience for its visitors and customers.

Continuous transactions monitoring ensures high functional availability for the application. You can read more Monitis blog posts about transactions monitoring best practices! Or Download Application and Transaction Monitoring Users Guide here.

To start using EC2 monitoring you can either sign up for a trial plan, which includes 2 EC2 instance monitoring possibility for 15 days, sign up for Plus plan in case you will need only 1 EC2 instance monitoring or build your plan yourself in order to have some flexibility in case the number of your instances is increasing or decreasing. After you have finished with the above mentioned steps, just login using your credentials and follow these steps to add EC2 monitor:

Step1: In this step you’re providing your Amazon account credentials. Please make sure that your private key name is exactly the same as the one you started your instance with. 

Step 2: In this step you’re going to add your contact(e-mail, IM or SMS) in order to be able to receive alerts related to your EC2 instances. You can skip timezone, send weekly reports, and contact group fields for now and change them later.

Step 3: In this step you will setup notification rules for your currently running instances, more clearly you will specify the max number of running instances allowed, your contact which we’ve added in the previous step, so you will get alert when the number of your currently running instances reaches your specify threshold.

Step 4: In this step you will setup policies(different monitors) you want each of your instances be monitored for. Please make sure to provide correctly your private key in Step 1, in order our agent can be automatically installed on your instance.

Step 4 is finalizing EC2 monitor setup process. The following module will be added automatically to your dashboard.

Drive, CPU, Load Average, Memory columns are related to internal policies selection and in case you’ve selected any of these during Step 4,  but you still see “Activate” values for them that means that something is wrong with agent installation(most probably related to your provided private key). HTTP, PING, SSH are checking your instance(s) from our monitoring locations and don’t require anything to be installed on your instance. Other configuration related columns like Instance ID, Image ID, Public DNS, Launch Time, Availability Zones and Key Name are gathered and shown automatically.

That’s it, your EC2 instance monitoring is up and running.

Provisioning both external availability and internal systems performance monitoring is one of key differentiators for Monitis. For internal monitoring Monitis provides downloadable agents which could be installed anywhere within client network. They will submit data to Monitis Central Server for storage, access and notification. When installed within corporate firewall, agents will communicate with the central server for user authentication and will check which tasks were assigned to them. Agents will use HTTPS so no need to change anything on firewall. Users (e.g. IT specialist like system or network administrators) can see a list of installed agents on Monitis web dashboards, and can remotely monitor server, network or intranet applications. Agent polls the server on regular intervals and when got a new monitoring task they start doing new probes. The Agent performance data transmitted over the internet to the Central Server can be watched in real time. The overall process of setting the agent and receiving performance metrics takes a few minutes. In contrast to software approach, with hosted service IT specialists don’t need to worry about having a dedicated server, database, notifications setup etc.

Agents are available for Windows, Linux and Sun Solaris platform. Agents can be installed on any user desktop and will not consume much CPU, memory and bandwidth and can co-exist with other applications. When installed, they can not only monitor the desktop or server they installed, but also any device in their neighborhood. Agents use ping, http, SNMP and WMI for monitoring intranets and they can also deliver CPU, memory, process and disk usage metrics.

Here are the IPs we monitor from:

US – 208.76.244.82
DE – 80.190.241.157
UK – 213.165.245.114
Panama – 190.5.236.105
Australia – currently there are some changes so no fixed IP at the moment just open your firewall, or exclude from the logs the following range 203.57.0.0/16