Oracle abandons commercial support for Glassfish JEE Server

Oracle abandons commercial support for Glassfish JEE Server

Summary: Oracle will no longer be supporting Glassfish Java Enterprise Edition Server business users.


When Oracle bought Sun, many people worried that Oracle would stop supporting Sun's open-source programs. We were right. Oracle first abandoned OpenSolaris, then OpenOffice, and now Oracle's Glassfish Java Enterprise Edition (JEE) 7 server.

GlassFish Logo

Officially, Oracle is still developing Glassfish Indeed, GlassFish Server Open Source Edition 4.1 is still scheduled for a 2014 release and its trunk code will eventually become the basis of GlassFish Server Open Source Edition 5 as a Java EE 8 implementation. In turn, Oracle continues, "the Java EE 8 Reference Implementation will be derived from GlassFish Server Open Source Edition 5."

But, and it's a big but, Oracle also said that it would "no longer release future major releases of Oracle GlassFish Server with commercial support — specifically Oracle GlassFish Server 4.x with commercial Java EE 7 support will not be released." Instead, "Oracle recommends that existing commercial Oracle GlassFish Server customers begin planning to move to Oracle WebLogic Server."

If your company is currently using GlassFish 3.1 with Oracle support, the clock is ticking. Premier support ends on March 2016 and extended support comes to its conclusion on March 2019. (PDF Link)

Oracle concluded, "GlassFish Server Open Source Edition continues to be the strategic foundation for Java EE reference implementation going forward. And for developers, updates will be delivered as needed to continue to deliver a great developer experience for GlassFish Server Open Source Edition."

That sounds good, but Java developers aren't buying it. Markus Eisele, a principal technology consultant working for msg systems ag in Germany, summarized Oracle's announcement as, "basically [being] about one thing: GlassFish Server as we know it today is deprecated from a full blown product to a toy product."

Without commercial support, Eisele sees Glassfish becoming less and less relevant to Oracle's customers and thus to Oracle and its WebLogic Server as well. Looking ahead, Eisele only sees one open-source JEE Server that he'll be able to recommend to business customers: Red Hat's WildFly, formerly known as JBoss Application Server.

Related Stories:

Topics: Enterprise Software, Open Source, Oracle, Servers

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Oracle "abandoned" neither OpenSolaris nor OpenOffice

    The OpenSolaris devs split because Larry Ellision wasn't giving the project the attention that they believed it deserved. OpenOffice devs were unhappy with Sun long before the Oracle acquisition, didn't see the improvement that they had hoped for with Oracle and forked it to LibreOffice. Whereupon Oracle donated OpenOffice to The Apache Software Foundation upon the recommendation of IBM.

    And how about some financials, Steven? How much money was Oracle making from OpenOffice and Glassfish? @TNOT, where are you? Your MBA perspective is required ...
    Rabid Howler Monkey
    • Looks like they weren't making enough to justify the expenses

      The financial model on Glassfish seems to be the Open Source Support Model, which is notoriously difficult to make money on.

      As Oracle is in business to make money, and Larry is pretty sticky about making money, I can see why they dumped Glassfish.

      It's a tough world out there, in business, make money or die. Glassfish died.

      Is that enough of a business analysis for you?
      The Nasty Old Troll
      • It was perfect!

        Clearly, the time you spent acquiring your MBA was well spent.

        Thanks, Cynical99 ... err, TNOT.
        Rabid Howler Monkey
  • Elli$on's greed kills glassfish

    this member of the axis of evil software is sticking it to the FOSS folks!
    LlNUX Geek
    • This shows a vulnerability in FOSS software

      This simply shows a vulnerability of FOSS projects that depend on large companies for support. When the winds of profit shift, the support may very well shift as well.

      In the case of Glassfish, it seems that the support monies earned from the project do not warrant future expenditures.

      As for Open Office and OpenSolaris, one could make the same assumptions. New management and new demands for profit may quickly doom any project in either the acquiring or the acquired company.

      Perhaps the lesson learned here is that FOSS projects should NOT depend on corporate monies to survive. Doing so has great risks as the demand for profit in companies seems to trump all,

      As a stockholder in various companies, I'd hang the CEO that lost money on a pet project that had little or no potential for profit. After all, the CEO is the servant of the stockholder, not the other way around. Now, getting the CEOs to recognize that can be a challenge.
      • Cynical99: "FOSS projects should NOT depend on corporate monies to survive"

        I would modify this to read that FOSS projects should be wary of depending on money from a single corporation to survive. The more corporations that support an open source project (think of the Linux kernel project), the less impact there will be when one of them bails.

        P.S. Since Steven is highly critical of Oracle's management of Sun's open source projects, one should also consider how the other potential suitor for Sun, IBM, would have managed them. With IBM, Solaris would be on its way out and I doubt if IBM would have funded OpenSolaris. In addition, I suspect that IBM would have donated OpenOffice to The Apache Software Foundation as Oracle did. How would the OpenOffice devs reacted to the Apache 2.0 license? This, in and of itself, might have caused the LibreOffice fork.
        Rabid Howler Monkey
        • Single corporation support -

          In this case, I would agree that a FOSS project depending upon a single corporation is at a fairly high risk, unless of course it makes enough money to cover all costs and a hefty profit besides.

          Glassfish seems to have a model that made some money, just not enough. OpenOffice never had money making models at all. They are free to the public, as in Beer. No valid attempt to recover any of the costs was ever made, so for Oracle to first ignore and then dump is quite reasonable.

          Libre fork may have been inevitable due to the nature of the SUN to Oracle transition. No disagreement there. Even if IBM had purchased SUN, IBM already had competing products in the past and had no real interest in Open Office. I'd say that Open Office as a corporate supported product never had much of a chance outside SUN's willingness to dump money into it. Once SUN was gone, so was the project.

          If you look at products supported by multiple companies, you'll note that the supporting companies have found ways to make enough to cover costs plus a tidy little profit from the product (Linux for example). You are correct that if one company supporting Linux bails, the impact is minimal.

          Point here is that an Office Suite distributed freely can't make money like Linux can once modified for a specific environment. At least that's IBM's story. Since they don't admit how much they make from Linux, we can't be sure how much money it brings in. That's a number I'd like to see.

          Having said all that, I must wonder why connecting TNOT and myself is of such importance to you. Obsessing about something so minor is, well, unhealthy. Stay healthy, stop obsessing. You'll lead a happier life.
          • Oh, and I forgot the Apache end run

            One must question Oracle's motivation for donating OpenOffice to the Apache Foundation.

            I see a couple of possible motivations -
            1 - Genuine desire to see Open Office move forward (seems a bit ludicrous to me)
            2 - Just didn't know what else to do with the code and really didn't care (possibility)
            3 - just wanted to irritate the Open Source movement by donating the code so the available resources would be split between Libre and Open, dooming both to be so mediocre that neither would succeed in the end.

            I personally lean to believing #2, but in the end, #3 will probably be the final destination of both projects.
      • The project isn't dead

        paid support for the project is dead. There is a big difference.

        Companies can decide to continue to use GF in the future, but if they are using for mission critical purposes, then they will either have to grow their own internal expertise to deal with problems or look for an alternative product to host their enterprise apps.

        This seems to be more about streamlining support costs within Oracle. WebLogic is their "official" platform and they have a large number of staff supporting it, GF is a product that was "foisted" upon them, when they bought Sun and they have been supporting it. Having extra support personnel for GF means extra cost and if not that many companies are using GF, then the costs will be too high and the cost centre will make a loss, therefore they want to push customers to a companion product that makes them more money and they can reduce their support centre overheads at the same time. A double win for Oracle, a lose-lose for GF customers who don't want or need WebLogic and other products out of the Oracle catalogue.
        • There's another option for Glass Fish Community Edition

          wright_is wrote:
          "they will either have to grow their own internal expertise to deal with problems or look for an alternative product to host their enterprise apps"

          Oracle is notorious for their expensive product support costs. Given that open source Glass Fish will continue to be developed, it represents a potential opportunity for a 3rd party to provide support for Glass Fish Community Edition. After all, it's 3rd parties that provide support for CentOS.
          Rabid Howler Monkey
          • Yep

            that is the beauty of the open source model... The question is, will somebody step in?
  • Adopting a FOSS project that competes with your cash cow

    Yeah, that's bound to work out well, when an extravagantly for-profit company takes over control of a FOSS project that happens to be in competition with one of their core cash-cow products.

    FOSS projects absolutely CAN work in a corporate setting - but they rely on the corporations involved being willing to accept profit margins than are smaller than they would be if they were aggressively pushing a proprietary product. They then make it back in volume, by winning the market-share war, by virtue of having a product with a lower TCO.

    But that's a fragile proposition, because of the FUD-slinging that proprietary companies can engage in.

    Nothing new here.
  • LogiCoy Provides 24/7 GlassFish Support

    Just because Oracle has discontinued their support services does not mean that consumers need to discontinue using GlassFish. Further, there is no need for an official hand off by Oracle to another group. Open source projects can get supported by other vendors that have engineers that know the code. LogiCoy has been providing and will continue to provide commercial production (24x7)support for all of the open source versions of GlassFish V2.1.x, V3.x,... Our engineers are the original and developers and architects of GlassFish and GlassFish ESB. So, if there is a desire to not migrate to Weblogic or other closed source app servers and continue to run the existing or future open source versions of GlassFish in production environments with 24x7 support and up to 1 hour SLAs, you may contact We have many high profile customers that have been and are currently using our 24x7 commercial support programs for GlassFish and GlassFish ESB for the past 5 years. Please see
    Fred Aabedi