NGINX is a web server that can also act as a load balancer, reverse proxy, etc. Together between them, NGINX and Apache serve more than 50% of the traffic on the web.  Today many companies are running their NGINX Open Source or NGINX Plus webservers on the Amazon EC2 environment using either Amazon Linux, Red Hat Linux, and Ubuntu.

Everyone agrees that it is a best practice to monitor applications like NGINX on EC2 and respond to any systems irregularities quickly.  Users expect fast access and constant uptime for their applications.

Current choices for monitoring NGINX webservers on Amazon EC2

Many companies are deploying Amazon CloudWatch to monitor their applications, and are even creating some levels of automation by developing scripts or by using AWS Lambda.  But configuring Amazon CloudWatch properly with custom metrics and setting up Amazon Lambda requires a certain amount of technical expertise that may be beyond that of many companies.  And then there is a cost and effort required to maintain any scripts as the applications evolve.

Another choice is to deploy an application performance monitoring (APM) solution, such as one from New Relic, Dynatrace, Datadog, or LogicMonitor.  APM solutions are great.  They do a really good job of watching over all your systems and pinpointing what happened and why.  They create logs that can be shared with and interpreted by your development team to recreate the issue and ensure that it doesn’t happen again.  But here’s the thing:  APM solutions provide a lot of data that you have to sort through (separating “signals from the noise”) and they do nothing to recover from failures when they occur.  APM tools are only part of the solution when it comes to reducing downtime for your NGINX webservers.

But some companies don’t have the internal staff or tools to monitor their EC2 environment themselves. This is the reason why they choose to outsource the task to a managed service provider.  There are some very real benefits to working with an MSP to manage your environment, such as not having to hire more staff as your environment expands, or not having to train your team on new technologies.  And the MSPs enjoy efficiencies as they can spread out their investments over many clients.  But there are downsides.  In some cases, you can be locked into high, fixed-cost contracts, and costs can escalate if issues are experienced and they have to escalate to address them.  And you lose continuity between the team that is monitoring the environments and those responsible for building and deploying the applications.

Whether you chose to invest in an APM solution or to outsource to an MSP, you still need to think about how quickly you can recover your NGINX webservers from downtime if and when it occurs.  We’d like to propose another alternative:  automated remediation with SOIS AppKeeper.

SIOS AppKeeper:  Automated remediation for NGINX webservers on EC2

Many of our customers have chosen to use SIOS AppKeeper to protect their NGINX webservers.  While they could have chosen a standard application performance monitoring (APM) solution or third-party monitoring solutions, they chose instead to rely on AppKeeper to automatically recover services or the entire EC2 instances if a failure occurs.  We will take a look at some of the reasons why and share with you a short video showing how AppKeeper works with NGINX.

SIOS AppKeeper is a SaaS service that is easy to install and configure and monitors any applications running on Amazon EC2, such as your NGINX webservers and their “nginx”, “cache manager”, and “worker” services.  When an anomaly is detected, AppKeeper automatically restarts the service, and if that doesn’t work it reboots the entire instance.  No more reading through painful logs to pinpoint the reason for the failure, or escalation to developers to restart your service or expensive outsourcing fees.  AppKeeper provides “set-it-and-forget-it” functionality so that you can rest assured knowing that your NGINX webservers are following EC2 monitoring best practices and are running properly, or will be quickly restarted if they experience any issues.

Today hundreds of companies rely on AppKeeper to keep their cloud environments running.  We invite you to check out this quick video for a demonstration of how AppKeeper protects NGINX webservers.

If you would like to try SIOS AppKeeper for yourself, we offer a 14-day free trial.  Simply click here to sign up.

Founded in 1932, Tokyo Diamond Tools Mfg. Co., Ltd. manufactures diamond tools for cutting, slicing, polishing, and drilling processes in a wide range of fields such as home electronics, semiconductors, electronic devices, healthcare and civil engineering. Although it is a long-established company with over 80 years of history, Tokyo Diamond has always been aggressively introducing new IT tools.  The company decided to move to Amazon AWS and virtualization for speed and efficiency of business improvements. SIOS AppKeeper played an important role in providing the application availability protection they needed. One of the reasons for Tokyo Diamond Tools’  move to the cloud and AWS was the Great East Japan Earthquake of 2011.  Although there was no direct damage to the servers at the head office, equipment at the Sendai plant in the Miyagi Prefecture dropped out of the rack, causing serious damage.  Tokyo Diamond saw the need for a  better business continuity plan.  This was a primary driver for the shift to virtualizing their core systems and using cloud environments.

At first, the company started migrating relatively small applications to the cloud.  “We started the move around November 2011.  At that time, we had no choice but AWS,” said Mr. Takuji Kokubo, the head of IT Systems at Tokyo Diamond.  They managed operations themselves using Sony Network Communications’ “Managed Cloud with AWS” cloud portal.  Mr. Kokubo, is a self-described “One Man IT operation,” so efficiency and automation are important to enable the smooth operations of the company’s IT systems, at locations in Japan as well as in Singapore and Thailand.

The move to the cloud and identifying the need for a recovery solution 

Diamond Tools implemented Amazon EC2 and Amazon S3, a cloud storage service. They moved their groupware, sales support and video conferencing systems to AWS.  Diamond Tools’ video conferencing system is critical to their daily operations.  “It is a tool that is very frequently used in various conferences and meetings with overseas locations.  It is used about 100 times a month, and with the transition from a conventional pay-as-you-go SaaS system to AWS, we could reduce the cost as much as two million yen a year,” said Mr. Kokubo.  Previously, voice quality often dropped when too many users including other companies connected to the system at the same time.  Then Mr. Kokubo would receive a complaint from users but the problem could not be solved easily.  Building a company-specific video conferencing system on AWS has stabilized video and audio quality and reduced complaints.

Mr. Kokubo soon experienced issues with their EC2 environment. “EC2 is stable as an infrastructure; however, sometimes services on it fail. I was always worried, and one day, during an overseas business trip, I got a call saying that users could not access the system.  I had to open my laptop, which I carry all the time, to connect to AWS via VPN and recover the failed service,” said Mr. Kokubo.  Clearly this process was not scalable.

Service failures frequently occurred on the instance where the groupware application ran.  Most of the company employees use the groupware application every morning to check the calendar or book meeting rooms. “Many people would call me  to ask what was going on if they experienced issues,” said Mr. Kokubo. He learned the risks of taking care of the system by himself through this experience.

“Ten years ago, systems would stop when there was a failure, and users were used to that. Today, however, I get complaints as soon as anything stops. Now, systems are expected to be up and running all the time, like water that comes out of a tap anytime.  System downtime is becoming more and more unacceptable,” said Mr. Kokubo.

Mr. Kokubo always felt that he had to take care of any issue when one of the services running on EC2 failed.  Being the only person able to address any issues continued to be a burden for him.  Mr. Kokubo said, “Being a one man IT shop, I tend to consider that my value lies in only providing a help desk function.  But it’s not good – I needed to consider how I can make the systems of Tokyo Diamond work without my support.”

Because he was the only IT professional at a company that was relying more on cloud services, he knew he needed help. “When Sony Network Communications told me that SIOS AppKeeper provides automated operations and management of EC2 instances,  I decided to use it even before looking into the details. If we could have a solution that recovers EC2 services automatically, then I wouldn’t need to troubleshoot every issue by connecting to VPN on-the-go,” Mr. Kokubo added.

SIOS AppKeeper is a cloud service that monitors EC2 instances and restarts services automatically when it detects any system impairments.  When a monitored service fails SIOS AppKeeper recovers the service automatically without having to have personnel intervene manually.

More than 10 Applications Running on AWS

Tokyo Diamond has continued to migrate other applications to AWS, and by June 2018 had more than 10 applications running on AWS. “I feel that AWS is useful: the operating system starts in 10 minutes and resources can be scaled up/down flexibly depending on the business. We can even delete a resource even if it does not work.  However, the core system that processes a large volume of data was migrated to a virtual environment using a data center, not AWS.  We are using both AWS and a virtual data center to build a system depending on the usage and needs,” said Mr. Kokubo.

Tokyo Diamond currently monitors three critical applications that employees rely on with AppKeeper, including their groupware application, sales force automation and their video conferencing system.  Mr. Kokubo said, “First, we classified our applications and decided to start with three of them. It is worth paying a fee if I can be freed from the daily mental burden and get peace of mind during a business trip.”

SIOS AppKeeper Enables IT Personnel to Focus on Other Tasks

According to Mr. Kokubo, he no longer needs to think about the operations and management of the instances monitored by AppKeeper.  AppKeeper is working perfectly.  It has freed up his time to focus on the virtual machines and their other AWS applications.  Mr. Kokubo feels that AppKeeper brings him a sense of security about these core applications.

However, he does have a request for the SIOS Technology development team about AppKeeper.  “If there are no events, I’m worried about whether it’s working properly, or I may forget how to login. It would be better if it notified me frequently that it is working,” said Mr. Kokubo.  Automatic operations and management are the strength of SIOS AppKeeper; however, it can be invisible when nothing happens.

Having said that, Mr. Kokubo agrees that it is very beneficial that he is freed from having to troubleshoot these applications.  Now he can spend more time on management, including evolving Diamond Tool’s IT strategy and planning, security, and BCP initiatives.

Although today Diamond Tool uses SIOS AppKeeper for the operations and management of three applications, it plans to extend AppKeeper’s coverage to other applications soon.  This will enable Mr. Kokubo to spend much time on those value-added activities.  “Even if we try to hire system personnel, it is difficult because Japan’s workforce is shrinking.  In the future, I expect system tools to replace people using AI (Artificial Intelligence), and I hope SIOS AppKeeper also develops AI-like functions to automatically address high-impact failures,” said Mr. Kokubo, hoping for a day when further automation of operations management will be realized.

Learn more about SIOS AppKeeper

Sign up for a Free Trial of SIOS AppKeeper