Sorry Ian - I reached out on gitter and did not see you online. The commit
has been made here:

*
https://github.com/geoserver/geoserver/commit/1d1bf02b97aafaadd15d4f7b69fa98011a7eb068


It picks up one back port from Andrea.

I also commented on
https://github.com/geoserver/geoserver/commit/96c501fc05de78c0e36ca2b0cc538f071ae4bd90

The comment "# deploy released community modules" was expected and
described the following:

   pushd community > /dev/null
  set +e
  mvn clean install deploy -P communityRelease -DskipTests $MAVEN_FLAGS
   set -e
   popd > /dev/null

So the real question is why set +e is not ignoring any failures when
deploying the community extensions?
--
Jody Garnett


On Mon, 22 Jul 2019 at 12:32, Ian Turton <[email protected]> wrote:

>
>
> On Mon, 22 Jul 2019 at 20:03, Torben Barsballe <
> [email protected]> wrote:
>
>> Hi Ian
>>
>> Note that while you can edit the release scripts on master and have them
>> affect any release, any changes to the pom must be in the commit you are
>> releasing, so you'll also need to rerun the geoserver-release job with the
>> new commit id before doing the publish.
>>
>> Also note that while gs-script-py is missing on the 2.14.x branch, it is
>> working fine on master and 2.15.x, and should not be removed from those
>> branches. The community modules build and were deployed without issue when
>> I did the 2.15.2 release last month.
>>
>
> Thanks, that would explain why it worked locally, I'll update 2.14 and
> restart the process
>
> Ian
>
>
_______________________________________________
Geoserver-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geoserver-devel

Reply via email to