06 Feb SQL Server 2017 & 2019 End of Life: The Impact of Not Upgrading

A lot of environments we see still run SQL Server 2017 or 2019 and sometimes even older, because hey, why upgrade? It still works, right? And indeed, it does, however the real impact starts when you look at support status and what changes once a version leaves mainstream support. 

Understanding Microsoft’s SQL Server Lifecycle 

SQL Server follows Microsoft’s lifecycle: a period of mainstream support followed by extended support. Mainstream support includes functional, performance, scalability, and security updates. Extended support includes security updates only. 

For SQL Server, this means that when it’s still under mainstream support, new functionality can be added, like additions and improvements to new features. For extended support, this means that when a security issue has been found, it will be fixed, but there won’t be any functional improvements anymore. 

The following table gives information about the SQL Server versions you might have and their end of support dates. 

 You see that there are just 2 versions which are in mainstream support and SQL Server 2017 is almost out of support. 

If you stay on SQL Server 2017, 2019, or older, the platform does not suddenly stop working. The impact is more subtle, and it usually shows up in risk, friction, and cost. Let’s check some of the reasons. 

You Lose Database Improvements That Make Life Easier 

Mainstream support is where real platform improvements happen. New features and refinements that help with stability, performance, and manageability. If you stay on a SQL Server version instead of upgrading, you are choosing to miss out on those improvements. 

A few examples from SQL Server 2025: 

• Optimised locking: designed to reduce lock memory usage, avoid lock escalation, and improve concurrency by holding far fewer locks even for large transactions. 
• Native JSON data type: JSON becomes a first-class data type instead of “just NVARCHAR with functions”. That improves validation and support for storing and working with JSON. 
• Vector data type and vector functions: Vector support enables storing and querying embeddings directly in SQL Server, which is the foundation for similarity search and AI-style retrieval scenarios. 

Even if it keeps running, you are locking yourself into older behaviour, older limitations, and more custom-made workarounds for problems newer versions support out of the box. 

Database Incidents Become Harder to Resolve 

Over time you might hit an edge case. A performance regression, a strange concurrency issue, a memory pressure problem, an availability scenario that behaves “just a bit different” than expected. On older versions, the options are often limited: 

• Less room for true fixes and more reliance on mitigations and workarounds 
• Longer root cause analysis because the platform is missing later quality improvements 
• Higher chance that you end up compensating at the application or infrastructure layer 

This is where “it still works” changes into “it works, but it costs you more time every time something goes wrong”. And there is another practical impact: Microsoft support gets more constrained as you move out of support. If your SQL Server version is out of extended support, Microsoft no longer provides security updates and you should expect limited to no product support for issues on that version. In practice that means you can end up in a situation where: 

• A support case turns into “upgrade first” before deeper troubleshooting will happen 
• You have fewer options when a bug or critical issue is confirmed 
• You are forced into riskier mitigations because the proper fix is simply not available anymore 

Audit and Governance Pressure Increases 

With the current NIS2 regulations, security breaches and risks will trigger questions when you are getting audited. 

• Why are we still on this version? 
• What is the risk acceptance rationale? 
• Which controls are in place to reduce exposure? 
• What is the long-term direction? 

If you do not upgrade, you spend time producing documentation and exceptions instead of spending that time on improvements. 

You Pay More, But You Get Less 

This one especially matters for those with “Software Assurance” bought as upgrades are included. Not upgrading increases: 

• Time spent troubleshooting and firefighting 
• Time spent validating compatibility across the stack 
• Time spent writing exceptions, policies, and risk acceptance documents 
• Time spent maintaining compensating controls 

At the same time, you get less value back from the platform because you are not benefiting from years of improvements. 

Planning Your SQL Server Migration 

If you have a SQL Server which is (almost) out of support, the best next step is not a rushed project. It is a short investigation that turns into a plan to migrate the instances. That investigation is where most upgrades succeed or fail. Not during the final cutover, but in the preparation that prevents surprises. 

Need Help with Your Database Migration? 

Monin specialises in SQL Server migrations, database performance optimisation, and ongoing database management. Our team helps organisations navigate end-of-life transitions smoothly, ensuring minimal downtime and maximum data integrity. 

Services we provide: 

• SQL Server migration planning and execution 
• Database performance tuning and optimisation 
• Database health assessments and upgrade readiness analysis 
• 24/7 database monitoring and support 
• Compliance and security consulting 

Contact Monin today to discuss your SQL Server upgrade strategy and ensure your databases are secure, supported, and performing at their best.