GitLab vs. Microsoft Defender for Cloud Apps vs. SonarQube

GitLab is good if you work a lot with code and do complex repository actions. It gives you a very good overview of what were the states of your branches and the files in them at different stages in time. It's also way easier and more efficient to write pipelines for CI\CD. It's easier to read and it's easier to write them. It takes fewer clicks to achieve the same things with GitLab than it does for competitor products.

Incentivized

Microsoft Defender for Cloud Apps is well suited when working with other Microsoft Applications. For example, if you are working with Microsoft Office 365 it works very well when implementing CASB features. It works when implementing monitoring or blocks on Sanctioned applications however customizing the message to users is not that great.

Incentivized

SonarQube is excellent if you start using it at the beginning when developing a new system, in this situation you will be able to fix things before they become spread and expensive to correct. It’s a bit less suitable to use on existing code with bad design as it’s usually too expensive to fix everything and only allows you to ensure the situation doesn’t get worse.

Incentivized

• GitLab excels in managing code versions, allowing easy tracking of changes, branch management, and merging contributions.
• It helps maintain code stability and reliability, saving time and effort in the development or research workflow.
• Powerful code review features, enabling collaboration and feedback among team members.
• Robust project management features, including issue tracking, kanban boards, and milestones.

Incentivized

• The integration to Microsoft Entra ID is seamless, which allows Conditional Access to redirect the session to Microsoft Defender for Cloud App for it to take actions (Block or Monitor).
• Tracker users' activity is very good when troubleshooting or running an investigate.
• Detecting risky users through tight integration with Microsoft Entra ID is a very good feature.
• Detecting mass downloads and blocking the download of files from non-manage company devices is a very good feature as well.

Incentivized

• Detecting bugs and vulnerabilities: SonarQube can identify a wide range of bugs and vulnerabilities in code, such as null pointer exceptions, SQL injection, and cross-site scripting (XSS) attacks. It uses static analysis to analyze the code and identify potential issues, and it can also integrate with dynamic analysis tools to provide even more detailed analysis.
• Measuring code quality: SonarQube can measure a wide range of code quality metrics, such as cyclomatic complexity, duplicated code, and code coverage. This can help teams understand the quality of their code and identify areas that need improvement.
• Providing actionable insights: SonarQube provides detailed information about issues in the code, including the file and line number where the issue occurs and the severity of the issue. This makes it easy for developers to understand and address issues in the code.
• Integrating with other tools: SonarQube can be integrated with a wide range of development tools and programming languages, such as Git, Maven, and Java. This allows teams to use SonarQube in their existing development workflow and take advantage of its powerful code analysis capabilities.
• Managing technical debt: SonarQube provides metrics and insights on the technical debt on the codebase, enabling teams to better prioritize issues to improve the quality of the code.
• Compliance with coding standards: SonarQube can check the code against industry standards like OWASP, CWE and more, making sure the code is compliant with security and coding standards.

Incentivized

• CI variables management is sometimes hard to use, for example, with File type variables. The scope of each variable is also hard to guess.
• Access Token: there are too many types (Personal, Project, global..), and it is hard to identify the scope and where it comes from once created.
• Runners: auto-scaled runners are for the moment hard to put in place, and monitoring is not easy.

Incentivized

• It takes some time to scan and apply the policies when there is some sensitive information.
• After it applies the policies, it works, but there is a delay.
• It doesn't provide any way to scan Microsoft Teams when an external exchange of images is happening. You can always do the filtering on the documents during the chat, but if there is an image, then some kind of OCR capability is required to detect it. At present, there is no way [Microsoft Cloud App Security] can go and detect those kinds of images and alert us

Incentivized

• Importing a new custom quality profile on SonarQube is a bit tricky, it can be made easier
• Every second time when we want to rerun the server, we have to restart the whole system, otherwise, the server stops and closes automatically
• When we generate a new report a second time and try to access the report, it shows details of the old report only and takes a lot of time to get updated with the details of the new and fresh report generated

Incentivized

I really feel the platform has matured quite faster than others, and it is always at the top of its game compared to the different vendors like GitHub, Azure pipelines, CircleCI, Travis, Jenkins. Since it provides, agents, CI/CD, repository hosting, Secrets management, user management, and Single Sign on; among other features

Incentivized

I find it easy to use, I haven't had to do the integration work, so that's why it is a 9/10, cause I can't speak to how easy that part was or the initial set up, but day to day use is great!

Incentivized

The interface is pretty simple and easy to use; however, you will need to do a lot of investigative research on your own to get comfortable with it. Originally, many of the Microsoft security tools had their own seperate consoles. Overtime, they have blended into one interface which is the ideal state. In some cases it is clear Microsoft had to pick which console a certain feature or setting was going to reside in and this leads to some confusion. For example, DLP is managed through Defender for Cloud Apps but you will also need to jump into Purview. For things like reverse proxy on your M365 tenant, you will need to go into Azure and setup conditional access rules. Not a big problem and I can understand why the settings are located where they are but for someone just starting out with Defender for Cloud Apps, it will take some time to figure out.

Incentivized

It can improve in some user experience and usability parts, like the code view and the way we assign issues it's a bit hidden and not highlighted

Incentivized

I've never had experienced outages from GItlab itself, but regarding the code I have deployed to Gitlab, the history helps a lot to trace the cause of the issue or performing a rollback to go back to a working version

Incentivized

GItlab reponsiveness is amazing, has never left me IDLE. I've never had issues even with complex projects. I have not experienced any issues when integrating it with agents for example or SSO

Incentivized

At this point, I do not have much experience with Gitlab support as I have never had to engage them. They have documentation that is helpful, not quite as extensive as other documentation, but helpful nonetheless. They also seem to be relatively responsive on social media platforms (twitter) and really thrived when GitHub was acquired by Microsoft

Incentivized

I have not utilized actual support but the Sales and Product teams have been super helpful in moving our implementation forward and showing us the best practices.

Incentivized

We we easily able to integrate the SonarQube steps into our TFS process via the Microsoft Marektplace, we didn't have the need to call SonarQube support. We've used their online documentation and community forum if we ran into any issues.

Incentivized

Gitlab seems more cutting-edge than GitHub; however, its AI tools are not yet as mature as those of CoPilot. It feels like the next-generation product, so as we selected a tool for our startup, we decided to invest in the disruptor in the space. While there are fewer out-of-the-box templates for Gitlab, we have never discovered a lack of feature parity.

Incentivized

More flexible and more features with easy integration with cloud services like Microsoft Azure and other cloud services. Overall both gives similar features but we prefer Microsoft cloud app security due to its high threat detection rate. mostly we have been able to stop the threat in very very less time.

Incentivized

SonarQube is an open-source. It's a scalable product. The costs for this application, for the kind of job it does, are pretty descent. Pipeline scan is more secured in SonarQube. Its a very good tool and its support multiple languages. Its main core competency is of static code analysis and that is why SonarQube exists and it does it exceedingly well. The quality of scan on code convention, best practices, coding standards, unit test coverage etc makes them one of the best competent tool in the market

Incentivized

I think is very well designed, and like any VCS it works as intended

Incentivized

• GitLab cut down our spent on container, package and infrastructure registry
• Best thing is we can now have everything in single platform which cost effective too
• Quality of support is really good and they do have emergency support team as well which is great

Incentivized

• Cloud App Security saves us thousands of dollars finding and rectifying apps security issues
• Identity Security Posture helps the organization identity stay in shape, saving thousands of dollars on security consultations
• The cost of suffering a breach cannot be quantified, CAS helps minimize the chances of the attackers succeeding, with excellent historical logging for most operations

Incentivized

• Positive ROI from the standpoint of flagging several issues that would have otherwise likely been unaddressed and caused more time to be spent closer to launch
• Slightly positive ROI from time-saving perspective (it's an automated check which is nice, but depending on the issues it finds, can take developers time to investigate and resolve)