accumulo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Christopher <ctubb...@apache.org>
Subject Re: [VOTE] Apache Accumulo 1.6.2 RC3
Date Thu, 29 Jan 2015 00:18:06 GMT
On Wed, Jan 28, 2015 at 2:30 PM, Sean Busbey <busbey@cloudera.com> wrote:
[snip]

> So that looks fine. I have seen cases before where using the maven compiler
> plugin's -source -target options without the correct rt.jar file resulted
> in Java 6 JVM compatible class files that still referenced JRE classes that
> weren't available.
>
>
Right, my main concern was this kind of problem (which I think can be
resolved by setting bootstrap classpath during compile).


> Attempting to compile the source tarball with a Java 6 JDK should cause
> that to show up.

[snip]

That's what I'd hope also, but I think there are fringe cases that wouldn't
catch this: use of constants which differ in value between versions,
changes between interface/abstract class, and maybe a few other fringe
cases that wouldn't be caught at compile time, but could cause runtime
errors. (I'm no expert on this, though, which is why I phrased it as a
question initially.)


> (as an aside, I couldn't find us actually documenting anywhere in the user
> manual or README what java versions we support.)
>
>
Maybe it'd be good to document it somewhere, but the java version is
specified in the pom, and has been:
Java 6 or newer for Accumulo < 1.7
Java 7 or newer for Accumulo >= 1.7

FWIW, we don't really document any other compatible dependency versions
either, outside the pom.xml, but divergence from this I'd typically expect
a downstream package maintainer to deal with (except for the fact that many
people use the upstream binaries directly, and that's a valid support case
for our community). More FWIW: if we were using maven to generate a site,
this kind of documentation could be generated automatically.


> On Wed, Jan 28, 2015 at 11:25 AM, Christopher <ctubbsii@apache.org
> <https://mail.google.com/mail/?view=cm&fs=1&tf=1&to=ctubbsii@apache.org>>
> wrote:
>
> > Does it matter that this was built with Java 1.7.0_25? Is that going to
> > cause issues running in a 1.6 JRE?
> >
> >
> > --
> > Christopher L Tubbs II
> > http://gravatar.com/ctubbsii
> >
> > On Wed, Jan 28, 2015 at 2:38 AM, Corey Nolet <cjnolet@apache.org
> > <https://mail.google.com/mail/?view=cm&fs=1&tf=1&to=cjnolet@apache.org>>
> > wrote:
> >
> > >   Devs,
> > >
> > >     Please consider the following candidate for Apache Accumulo 1.6.2
> > >
> > >     Branch: 1.6.2-rc3
> > >     SHA1: 3a6987470c1e5090a2ca159614a80f0fa50393bf
> > >     Staging Repository:
> > >
> >
> https://repository.apache.org/content/repositories/orgapacheaccumulo-1021/
> > >
> > >     Source tarball:
> > >
> > >
> >
> https://repository.apache.org/content/repositories/orgapacheaccumulo-1021/org/apache/accumulo/accumulo/1.6.2/accumulo-1.6.2-src.tar.gz
> > >     Binary tarball:
> > >
> > >
> >
> https://repository.apache.org/content/repositories/orgapacheaccumulo-1021/org/apache/accumulo/accumulo/1.6.2/accumulo-1.6.2-bin.tar.gz
> > >     (Append ".sha1", ".md5" or ".asc" to download the signature/hash
> for
> > a
> > > given artifact.)
> > >
> > >     Signing keys available at:
> https://www.apache.org/dist/accumulo/KEYS
> > >
> > >     Over 1.6.1, we have 148 issues resolved:
> > >
> > >
> >
> https://git-wip-us.apache.org/repos/asf?p=accumulo.git;a=blob_plain;f=CHANGES;hb=1.6.2-rc3
> > >
> > >     Testing: All unit, integration and functional tests are passing.
> > >
> > >     API compatibility report for 1.6.1 to 1.6.2:
> > >
> > >
> >
> http://people.apache.org/~cjnolet/accumulo-1.6.2-rc3/compat_reports/accumulo/1.6.1_to_1.6.2/compat_report.html
> > >
> > >     API backwards compatibility report for 1.6.2 to 1.6.1:
> > >
> > >
> >
> http://people.apache.org/~cjnolet/accumulo-1.6.2-rc3/compat_reports/accumulo/1.6.2_to_1.6.1/compat_report.html
> > >
> > >     The vote will be open until Saturday, January 31st 12:00AM UTC
> (1/30
> > > 8:00PM ET, 1/30 5:00PM PT)
> > >
> >
>
>
>
> --
> Sean
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message