July 2026 – Effectively the end of the DirectAdmin Legacy License

[virtio]

July 2026 – Effectively the end of the DirectAdmin Legacy License​

For anyone who wasn’t aware yet: July 2026 effectively marks the end of the DirectAdmin Legacy License — at least, that’s the unavoidable conclusion after our recent discussions with DirectAdmin.

Over the past week, we had an in-depth exchange with the DA team about the future of the Legacy License. Since many providers/users appear unaware of what is coming, we want to share a clear and transparent summary to help others plan responsibly.

How this started​

The conversation was triggered by a simple fact:

• MariaDB 10.6 reaches community EOL in mid-2026.
• Legacy installations cannot upgrade to MariaDB 10.11+ via CustomBuild, because newer versions are restricted to the new DA licensing model.

In other words: when 10.6 hits EOL, Legacy systems can no longer run a supported MariaDB version inside the CustomBuild ecosystem, which is how the vast majority of DA users maintain their LAMP stack.

---

What DirectAdmin clarified​

After several messages back and forth, DA confirmed the following points clearly and unambiguously:

1. Legacy should now be treated as effectively hard EOL.​

Although previously described as “sunset” or “soft EOL,” DirectAdmin clarified that, in practice, Legacy is a static product.
No further development will be added — not for MariaDB, not for OS support, not for any LAMP components.

2. MariaDB 10.11+ will not be made available on Legacy — now or ever.​

This includes all maintained versions beyond 10.6.

3. DA acknowledges Legacy will become unusable for most production environments.​

Their own wording:

“We acknowledge this will make licenses useless for most, but that’s the intent behind lifecycle management/EOL.”

4. The only alternative is using OS-repo MariaDB/MySQL packages.​

However, these are typically older, behave differently, and are not aligned with how DA environments are normally managed via CustomBuild.

5. More “technical walls” are expected — by design.​

MariaDB is simply the first.
Newer OS versions (RHEL 10, Debian 13, etc.) are also unsupported and will remain incompatible with Legacy.

---

Why this matters​

For years, the industry understanding of the Legacy License was:

• still fully functional,
• without the newer features of the updated license model,
• and to be retired at a later, clearly announced EOL date.

Not as a license that would become effectively non-viable due to blocked updates of core security-critical components.
Many providers — including larger ISPs — were unaware that MariaDB 10.6 EOL forces a de facto end-of-life for Legacy by July 2026.

In operational reality, this means:

Legacy becomes unsuitable for commercial workloads from mid-2026 onward.
​

What this means for you​

If you’re still running DirectAdmin Legacy:

• Migration will be required — sooner than expected.
• Assume Legacy is already operationally EOL.
• New deployments on Legacy should be avoided for security and compliance reasons.

DirectAdmin did provide full clarity once pressed with direct questions — but only then. Given how many users still operate under outdated assumptions, sharing this information openly feels necessary.

Clearer communication would help the ecosystem​

A simple, explicit statement such as:

“Legacy is now effectively hard EOL. No further updates of core components such as MariaDB will be provided. Expect full deprecation by mid-2026.”

…would prevent many partners and users from being caught off guard.

Closing – and a very important note​

This post is not meant as criticism of DirectAdmin’s decision to evolve its licensing model. A product lifecycle is normal, and DA is absolutely entitled to modernise.

More importantly:

We continue to be strong supporters of DirectAdmin.​

We still consider DA the best control panel in the market - lightweight, stable, fast, and with a far superior operational experience compared to many alternatives. Nothing about this situation changes the fact that DA remains our preferred panel, and the one we recommend to customers.

Our comments are purely about clarity:

the first hard technical wall (MariaDB) accelerates the practical end-of-life of Legacy much earlier than many expected.

If this summary helps you prepare your infrastructure or update your customer communication, then it has served its purpose. Also, please check the poll.

 

[Active8]

MariaDB 10.6 does not mean directly end of Legacy DA, there are still server that are running MySQL 5 without problems as long you don't have open your SQL to the world you can work with 10.6 for al least few years,
Yes no updates but I am sure that many software will still support 10.6.
I think that most of the people will migrate after OS is EOL and not MySQL

 

[Richard G]

This includes all maintained versions beyond 10.6.

Correct, however, there might be a slight chance that Directadmin would provide an option so the OS can keep the MariaDB installation up to date.
Almalinux 9 has updates for MariaDB 10.5 until 2032 theoretically.
That would mean that if we would downgrade to MariaDB 10.5 the OS could provide us update without further development requirement of DA and it would still be up to date.

Maybe @romans can enlighten us if there might be plans providing this for us to have the MariaDB updates switch to the OS.

As for the rest of your article. That is already known for several years. So people have enough time to make changes or choices.

 

[virtio]

Thanks both for your replies — much appreciated.

@Active8

You’re absolutely right that running an older database version doesn’t mean a system immediately becomes unusable. Many environments continue to run fine for quite some time, especially when the database isn’t exposed publicly. Anyone is free to make that choice, and for some setups that will be perfectly acceptable.

Our point is mainly that the first unavoidable blocking point for many providers appears earlier than expected — not because MariaDB 10.6 stops working overnight, but because most hosting companies rely on supported, actively maintained database versions as part of their operational and security policies. For those providers, the practical EOL arrives as soon as they can no longer keep the LAMP stack aligned with their standards. That’s the angle from which this post was written.

@Richard G

The idea of using OS-maintained MariaDB packages is indeed a potential technical workaround, and if DA considers enabling such an option, that would be valuable information for many users. We’d gladly hear more from @romans if that’s something being explored. But it would still represent a shift away from the CustomBuild-based workflows that most providers currently depend on.

Regarding the point that “this has been known for years”: absolutely — but based on the conversations we’ve had with several ISPs in the past weeks, it seems clear that quite a few people were unaware of the practical, not just theoretical, implications. Hence this topic to make sure everyone has the same information and can plan accordingly.

To be clear:

Everyone is, of course, entirely free to approach this transition however they prefer, based on their own risk profile, business needs, and operational policies. This post is simply intended to share the insights we received directly from DA, so those who want to plan proactively can do so with full information.

As a final note, I’d like to add our broader view on the matter.

In our opinion, it would ultimately be healthier — both technically and strategically — for DirectAdmin to give the Legacy edition a clear and hard end-of-life date. Legacy is deprecated software, and history shows that as long as something deprecated remains available, there will always be people (and even entire organisations) who continue relying on it far past what is wise or secure. A hard EOL gives clarity, protects users from themselves, and enables everyone to plan properly instead of delaying the inevitable.

A wise man once illustrated the concept of EOL sunsetting with this metaphor:

“EOL sunsetting is like a plane making a soft landing and giving everyone plenty of time to deboard. Some exit quickly, others take longer. But when someone asks, ‘If I stay, what do I get?’ - it misses the point. The goal is to deboard, not to plan for continued use after landing.”

I think that captures the situation perfectly.

Legacy served its purpose for many years, and it did so extremely well. But eventually, it has to be retired - and that’s ultimately better for DirectAdmin, for providers, and for end-users. Clear timelines and a firm EOL would help everyone move forward confidently.