RHEL 8 Speculation ???

Home » CentOS » RHEL 8 Speculation ???
CentOS 13 Comments

Is there any blog that has information on a potential RHEL 8 release date?

boost in 7 is now too old for some things, in addition to gcc. There are solutions in 7 to those issues but it’s starting to feel like 6 felt shortly before 7 came out, so I wonder if it is getting near to time.

I’m working on a major project bitcoin related and it would be frustrating to deploy a bunch of CentOS 7 virtual machines only to have
8 come out fairly soon afterwards.

13 thoughts on - RHEL 8 Speculation ???

  • I’d expect the release of RHEL 8 no less than 6 months after a beta was released, and I haven’t heard anything about a new beta release. It’s probably not going to happen real soon.

    Having said that, I didn’t realize until how close RHEL 7 is getting to the three-year mark. I wouldn’t be surprised to see a beta 6-12 months from now, and a new release in 12-18 months.

  • I have no real information on this either .. but one thing to think about is that RHEL-7 is much less conservative with ‘rebases’ than the older RHEL versions. (Case in point, major shifts in Gnome, KDE, Xorg, etc.)

    With containers and software collections, and with more aggressive rebases in the Desktop space, I see the timelines becoming a little longer between major versions.

    Also, as others have said, I see no alpha or beta for RHEL-8 anywhere. RHEL-5 does go EOL at the end of March 2017, so I guess a beta could happen soon(ish).

    • “I see no alpha or beta for RHEL-8 anywhere.”

      Since each major RHEL version is generally based on a particular version of Fedora, there is certainly no need for an alpha. So, the corresponding Fedora version will directly serve as entry into the beta phase!
      The release of Fedora 26 has been announced for June of this year (see https://fedoraproject.org/wiki/Releases/26/Schedule). I expected that roughly at this point Red Hat will start the beta phase of RHEL8, which will be based on that Fedora release.

  • We use GCC 6 from SoftwareCollection 6 and build our own Boost libraries with static linking. The result binaries work on all C7 instances just fine without the need to install any extra packages. The binary size isn’t bloated too much by statically linking Boost, because many Boost functions are header templates anyway (but your mileage may vary).

    Not sure if you have other dependencies but we never feel the need to upgrade the whole OS just to get a newer GCC and Boost.


  • Never been a fan of static linking. Exploit in a library and you have to rebuild everything that links against it.

    That’s one of the problems in the android world. Some library is found to be vulnerable, it gets fixed, but a lot of apps on the play store remain vulnerable simply because the vendor never submitted a rebuild.

    That’s also my beef with php composer, a lot of webapps lock the version of the library they use and composer happily grabs the crusty old version with its flaws.

    But anyway, not suggesting a whole new OS just because of gcc and boost. It’s a trend I’ve been seeing where recently, more and more apps refuse to build become one or more of the build dependencies is too old.

    There’s always a solution, but its nice when things just work.

  • “Never been a fan of static linking. Exploit in a library and you have to rebuild everything that links against it.”

    This is not a good reason for a lot (most) of Boost, which is template based anyway. You would have to rebuild anyway, even if you used dynamic linking.

  • Looking at the release interval of the last three major versions of RHEL (with three and a half year in between of each new release), I would expect RHEL 8 coming out at the end of this or beginning of next year – given the fact that RHEL7 has been released in June of 2014…
    So, I expect the first beta version of RHEL8 in summer of this year with a period of roughly another half year.

  • I don’t think enough has changed between Fedora 19 and Fedora 26 to make it interesting enough yet to base RHEL 8 off of. Rather, I think we will see more work to make Fedora 27 even more modular that will eventually be pushed as a key feature of RHEL 8. So, my guess is that Fedora 27 will be released around November (currently October is targeted) and then shortly afterward we will see a RHEL 8 beta.

    I also believe Fedora 27 would be the last Fedora release that RHEL 8 could be based off of. Red Hat has usually made sure there is always a RHEL which has been out for at least one year which is under phase 1 support in their product life-cycle. Since RHEL 7 phase 1 support ends Q4 of 2019, RHEL 8 should be released before or during Q4 2018. If there is a 6 month beta period, then RHEL 8 beta should be released around Q2 2018. Since Fedora 28 probably won’t come out until the end of Q2 2018, it is probably cutting the timeline too close to expect a Fedora 28 based RHEL 8.

  • “I don’t think enough has changed between Fedora 19 and Fedora 26 to make it interesting enough yet to base RHEL 8 off of.”

    Major kernel release: 3.9.5 -> 4.11 which adds a million features relevant to containers, GPU, other.
    Package manager change: yum -> dnf which provides major performance gains.

    What HASN’T changed from 19-26? So much that backwards compatibility of other parts off the RH stack potentially break and it will probably take a lot of engineering work to produce a stable 8 certified by hardware vendors which need to update their drivers.

  • As John Boero pointed out, I should have qualified my statements better. I didn’t see any killer feature introduced by Fedora 19 to Fedora 26. While I like a lot of what had already changed, I don’t see them as being the killer item to convince a CIO with. I was looking for a change that I would expect Red Hat sales to keep pointing to as the reason to use RHEL 8 instead of RHEL 7. I think 2019 will be the year of RH modular server sales webinars.

    As to RHEL 8 being released in 2019, I don’t think RH will do so since it would mean it would be released the same year RHEL 7 goes into phase 2 support. Instead, I still think RHEL 8 will be released Q4 2018. My estimate on the timeline also makes the timing tight for Fedora 28 and kernel 4.16 to make it in but might be possible if the RHEL 8 beta period is kept short. I do not expect that RHEL 8 based on Fedora 29 is likely to happen.