Apache Hadoop End of Life - Dates and Lifecycle

Enterprise data platforms built on Hadoop in the early 2010s are often still running it today, since it typically sits at the core of a company's broader data infrastructure and migrating off it is rarely a quick project, which is part of why version sprawl is common across real Hadoop deployments.

Apache Hadoop Lifecycle Explained

Hadoop's end of life process is driven entirely by volunteer availability, not a calendar. The project's own policy states plainly that a branch is declared end of life "if no volunteer to do a maintenance release in a short to mid-term (like 3 months to 1 or 1.5 year)," and even after that, the community can still choose to do a security-only release if someone steps up. This is why Hadoop currently has four simultaneously active branches, 2.10, 3.3, 3.4, and 3.5, rather than the usual pattern of retiring older lines once a new one ships: each stays open only as long as someone is actually willing to maintain it.