In January, OverOps announced that it is expanding its support to the .NET ecosystem.
OverOps is a continuous reliability platform that allows developers to analyze and optimize code at runtime.With OverOps, engineering teams can better identify when rapid code changes introduce critical issues and resolve them quickly to prevent a negative impact on customer experience.
An OverOps survey in 2017 showed that engineering teams were spending at least 25% of their time, on average, solving production issues.
InfoQ interviewed Tal Weiss, co-founder, CEO, and CTO at OverOps, about how engineers can benefit from using OverOps and the future of automating the production debugging process for .NET.
InfoQ: Why OverOps decided to support the .NET ecosystem?
Tal Weiss: The trend we’ve observed within the enterprise community is that there are usually two "sides of the house" when it comes to their mission critical applications: the Java side and the .NET side. With the addition of .NET compatibility, OverOps can now support the vast majority of financial, healthcare and media enterprise applications, helping our customers continuously deliver reliable software across their entire estate.
InfoQ: How can OverOps for .NET help organizations uncover and address code quality issues in live applications?
Weiss: With OverOps for .NET, users can continuously verify their code as it moves from the developers’ fingertips all the way to successfully serving customers in production. OverOps analyzes the execution of code across test, staging and production, providing application error analytics against a dynamically adjusted baseline. This enables OverOps to spot any regressions (such as new and increasing errors) in every release without having to rely on developer foresight.
InfoQ: Can OverOps help teams move their production services to .NET Core?
Weiss: Absolutely. Migrating a .NET application to Core may likely require a major change and refactor in what could be a large and complex codebase. This will most certainly introduce a significant degree of breakage, as unintended consequences are likely to occur as the result of large changes. OverOps enables teams to detect and resolve critical software issues stemming from changes the instant they manifest, without requiring developer foresight, to ensure an impact on users is minimal to none.
InfoQ: What is the most effective way to cut down on the time that teams spend debugging errors?
Weiss: There are two critical components in cutting down the MTTR of every software issue: Earlier detection within the SDLC; the sooner an issue is detected from the moment of its introduction, the easier it is to resolve, as the information about its creation is still "fresh" in the developers mind and it has less time to intertangle with other issues and create a cascading effect which will make it hard to pinpoint the original issue. Accurate context: the more information the developer has as to the exact state of the code at the point of execution, the faster it is to reproduce and resolve the issue. Dependence on limited logging information and the need to re-instrument code to add additional outputs can play a significant role in lengthening the time it takes to debug software issues.
InfoQ: How can an engineering team integrate OverOps within its daily workflow?
Weiss: OverOps seamlessly integrates into the software toolchain used to deliver software. This includes integrations into collaboration tools such as Slack, Jira, ServiceNow and even email to route issues to the right developers. Integrations into CI/CD platforms and static analysis such as Jenkins, SonarQube and Bamboo enable QA teams to verify releases and prevent critical errors from being pushed into higher level environments. Lastly, integrations into monitoring tools such as APMs and Log analyzers provide developers with much deeper context into the execution of code to enable much faster resolution of software issues and reduce the back-and-forth between developers and Ops.
InfoQ: Can OverOps replace the tester’s role in QA?
Weiss: OverOps cannot replace the tester, but it can help make the tester much more effective by reducing his/her reliance on manual developer foresight. By leveraging OverOps’ continuous runtime code analysis and machine learning, QA teams can spot a much wider array of issues that developers did not test for when writing the code, helping ensure that critical software issues do not slip by. When an issue is detected, OverOps provides QA and developers with a much richer context as to the exact state of the code, helping reduce both the time to debug and the likelihood of an issue becoming unreproducible.
InfoQ: What is planned for OverOps integration with Azure DevOps Services?
Weiss: We are working hard to deliver the same levels of deep integration we have with leading CI/CD platforms such as Jenkins into the Azure CI/CD pipelines to provide a fully native and seamless experience for Azure and .NET users.
InfoQ: What is the next step for OverOps for .NET?
Weiss: One of our next main objectives for .NET is delivering on native interactions into the Azure DevOps platform, and to support .NET Core for Linux to ensure our users can continuously deliver better .NET code no matter where and how they deploy and run their code.
InfoQ readers can find more information on the architecture page or request a demo to try OverOps for .NET.