Digital Media publishing platform approached StackExpress for a completely custom application support, and off hours incident support package. They required dedicated team of Engineers, a customized training plan, and required us to work in their very specific and evolving workflow. After extensive evaluation on some very stringent parameters, especially in the area of data and physical security, StackExpress was awarded the contract. The client provided an extensive documentation, a thorough on-boarding, and training plan. A team of our engineers went through multiple rounds of training with the client with performance evaluation at each step.
The client application is multi-tier, developed in C#, IIS for web server, and is backed by MS-SQL server for the database layer. The application is deployed on AWS, and uses regular set of most used AWS services like CloudFront, S3, lambda and RDS. They have a single tenant architecture due to business and legal reasons, and hence, have a higher overhead of deployments.
StackExpress team was initially given responsibility to shadow the existing team and then take over level 2 application support. The team executed it well enough that the client had let go local unit of well know multinational IT company handling their first level application support, and handed over the responsibility to StackExpress. Today we are handling all tiers of their application support, and also shares responsibility of build and release. The client has also opted for our off hours monitoring support, so their team can sleep in peace, knowing StackExpress is there to handle any incident
A New York based...
Web development team approached StackExpress for DevOps practices, and to manage their client application deployments on public cloud.
Their first project was to setup, and manage a MEAN stack application in fail-over configuration. We built an understanding of the application, and identified deployment needs. The application was deployed on Microsoft Azure, with multiple levels of monitoring using Microsoft Azure's built-in monitoring capabilities, and Datadog. Alerts and on-call schedules were managed using PagerDuty.
The configuration management and application deployment is managed using Ansible. A Master-slave setup for the database layer was employed to add further reliability to the deployment. The project was completed with automation of the deployment cycle using Jenkins. With a Continuous Integration (CI), and Continuous deployment (CD) in place, the development team is able to achieve faster, and stress free automated deployments. For further peace of mind, StackExpress shoulders responsibility of on-call rotation, and primary incident response.
Our 24x7 monitoring and incident response team handles the first response, and our team of DevOps Engineers investigates the incidents needing deeper involved like an Out of memory issues, or an anomaly in load balancer error rate metric.
A relevant DR and backup strategy for the database layer was also implemented. Client is now growing without having to worry about CI/CD and Production issues.
A Silicon Valley...
Startup with a team of 30+ developers contacted us. They had a free to use software platform with millions of users. They wanted StackExpress to look after their RoR/Docker platform, develop features, look after abuse issues and reduce running cost. This was necessary for them to focus on new service they were developing.
Everything was well documented, so it didn’t took us long to set up their staging. StackExpress studied their AWS running cost and came up with the plan to move everything over to cloud that our company runs. Our proposal would cut running cost by 2/3rd. Startup was spending tens of thousands of dollars, our proposal thus had significant savings. We got a go ahead from the founder.
In a 4 weeks time we moved complete app and data to our cloud without causing any downtime. In next few months we developed features to stop abuse. StackExpress implemented users forums and OAuth. Overall 99.9% up time of production was ensured.
A US startup...
Founded in 2013 contacted StackExpress in hurry. All in their development team have left the previous week and DevOps Engineer was leaving the same week. Startup has lots of clients.
There was no or very little documents available on their RoR, Salt App Stack. During 1 hour kick-off meeting, founder and DevOps Engineer explained App functionality and Stack. Thereafter no help was available from their Developers or DevOps Engineer.
We were giving access to their repository and stack. Our challenge was to build app and stack understanding fast to ensure app continue to serve their client.
In first week we mostly documents their app and stack working. Next few weeks we set up their staging. Luckily there were very little production issues initially which were solved by clearing log files etc.
During the 10th week suddenly application stopped working. Their customers requests were no processing and that had created big backlog. Customers were threatening to leave our client. This was the biggest challenge. Client was totally dependent on us and could help us very little on App working. We put together a team consisting of a DevOps Engineers and a Developer. Team worked almost 36 hours non-stop finding issues and fixing them down the line. Initially we just made App working by fine tuning stack but some delays in processing were still there, later we got down to code issue and fixed that too. Crises was averted and our client was able to retain all their customers.
In later years, client hired a new Developers team. We helped them setup their environment and explained app working. Our client is now focused on feature development and growing the business.