JDK 11.0.3 Update process
Andrew Hughes
gnu.andrew at redhat.com
Mon Feb 18 22:44:49 UTC 2019
On Mon, 18 Feb 2019 at 19:26, Langer, Christoph
<christoph.langer at sap.com> wrote:
>
> Hi,
>
> I agree with Aleksey's point of view on the backports. We're currently processing them and I believe it'll take a little bit of time until we're done. Today, for instance, I stumbled over a JFR change which needs some more thinking/testing. So, for the time being I think it's best we keep jdk11u open for that work. So hopefully some time next week we'll be done with the main bulk and can proceed going towards the new setup.
>
> I also think, we should get the answer to Volker's questions about the right way to do merges with these mercurial trees and the hgupdater tool before we install the new repository setup.
>
> Consider a point when we have the target setup in place (as per Rob's mail which fits perfectly by the way, thanks Rob):
> jdk11u -> 11.0.3 (Changes pushed to jdk11u will create a JBS backport issue targeted for version 11.0.3)
> jdk11u-dev -> 11.0.4 (Changes pushed to jdk11u-dev will create a JBS backport issue targeted for version 11.0.4)
>
> I would think we should then push changes targeting version 11.0.3 into the jdk11u repo. Then, by merging jdk11u regularly into jdk11u-dev, we'll get the 11.0.4 backport items created by hgupdater. Rob needs to confirm that. And what happens, if a change in jdk11u-dev needs to be cherry-picked to jdk11u? will hgupdater be smart enough to cope with that when merging back jdk11u into jdk11u-dev? And, what happens, if after April we set jdk11u to target 11.04 in hgupdater and then merge jdk11u-dev into jdk11u? Will hgupdater be smart enough to handle that? I believe/hope the answer to both questions is a yes, but I'd like to have Rob or somebody from the Oracle ops team confirm.
>
> We should also find an agreement on the question what the restrictions for jdk11u repository will be eventually? Shall it be all jdk-updates committers or only a subset of people performing the release work? I'm open here.
>
> And last but not least, Andrew (APH): As of today you are the appointed jdk11u maintainer [0]. Congratulations Now I think this means that you'll need to approve our "jdk11u-fix-request" labeling in JBS. And there are quite a few of them. I suggest you name a few additional maintainers like Oracle has for the jdk12u updates project [1], that have the powers to approve these fix requests. I'd like to hereby nominate Aleksey, Andrew (Hughes) and myself for this role. We should of course stick to Alekseys suggested rule that if one of us requests a backport, somebody else from the maintainers group has to approve to follow a 4-eyes principle. Would you agree to that?
>
> Thanks and Best regards
> Christoph
>
> [0] https://wiki.openjdk.java.net/display/JDKUpdates/JDK11u
> [1] https://wiki.openjdk.java.net/display/JDKUpdates/JDK12u
>
It does seem a bit early to me for 11.0.4, especially with the
transition from Oracle ownership. I was expecting
the branch to happen more around the time we received the security
patches, but then I also feel that the lead
time under Oracle was too long, leading to patches taking a long time
to actually appear in releases. It really
depends on what testing one wants to do prior to release.
Thank you for the nomination, by the way. Accepted :-)
--
Andrew :)
Senior Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)
Web Site: http://fuseyism.com
Twitter: https://twitter.com/gnu_andrew_java
PGP Key: ed25519/0xCFDA0F9B35964222 (hkp://keys.gnupg.net)
Fingerprint = 5132 579D D154 0ED2 3E04 C5A0 CFDA 0F9B 3596 4222
More information about the jdk-updates-dev
mailing list