p2 director install didn't complain, so no harm, no foul there.
install-grinder has been broken for a while now due to issues on the
slaves on which it runs.
do we have jira for that and followed up with ci-issues to get that
fixed ?
if we don't have install-grinder to verify we are back to having to
verify stuff tediously manually.
/max
On 06/10/2015 05:07 AM, Max Rydahl Andersen wrote:
> On 4 Jun 2015, at 14:58, Nick Boldt wrote:
>
>> Versionwatch CAN be tweaked to add exceptions. We currently have 3
>> in
>> there for know issues where multiple versions were needed.
>>
>> But it doesn't validate IF those exclusions are still true.
>
> yes so it does not have anything to do with version watch tests
> failing.
>
>> For that we need to build everything, install everything, and see
>> what's broken. Suspect that some of these IUs MIGHT be needed for
>> something from Central. I didn't test-install that.
>
> Are you saying you removed IU's we have but did not run the tests we
> already have in place to verify such changes ?
>
> p2install and install grinder ?
>
> We made those to make this kind of thing easier to test/verify.
>
> Thus lets use them.
>
> /max
>
>
>> N
>>
>> On 06/04/2015 07:13 AM, Max Rydahl Andersen wrote:
>>> On 4 Jun 2015, at 0:51, Nick Boldt wrote:
>>>
>>>> JBDS built locally and installed fine. Resulting IUs in
>>>> installation
>>>> (no
>>>> source bundles, as expected):
>>>>
>>>> javax.servlet.jsp_2.2.0.v201112011158.jar
>>>> javax.servlet_3.1.0.v201410161800.jar
>>>> org.apache.commons.lang_2.6.0.v201404270220.jar
>>>>
>>>> Can someone else verify this change is safe?
>>>>
>>>> Motivation here is to remove dupe IUs and therefore get a cleaner
>>>> install footprint
>>>
>>> if the separate versions are not required then that is a good
>>> reason.
>>>
>>>> , which will also make versionwatch builds blue again.
>>>
>>> I would rather fix version watch to be able to add exceptions for
>>> this rule
>>> since we most likely will end up with multiple versions of things -
>>> but
>>> that
>>> should only be marked okey if it is actually needed/wanted.
>>>
>>> I can't verify it easily today - maybe Denis can on his buildsetup
>>> ?
>>>
>>> /max
>>>
>>>> See also
https://issues.jboss.org/browse/JBIDE-19605
>>>>
>>>> On 06/03/2015 06:15 PM, Nick Boldt wrote:
>>>>> Currently building this locally and then will attempt to build
>>>>> JBDS
>>>>> with
>>>>> it:
>>>>>
>>>>>
https://github.com/jbosstools/jbosstools-target-platforms/pull/151
>>>>>
>>>>> Resulting TP contains only a single version of these IUs (instead
>>>>> of
>>>>> multiple versions as we did in the past):
>>>>>
>>>>> javax.servlet_3.1.0.v201410161800.jar
>>>>> javax.servlet.jsp.source_2.2.0.v201112011158.jar
>>>>> javax.servlet.source_3.1.0.v201410161800.jar
>>>>> org.apache.commons.lang_2.6.0.v201404270220.jar
>>>>> org.apache.commons.lang.source_2.6.0.v201404270220.jar
>>>>>
>>>>> To build against local TP:
>>>>>
>>>>> mvn clean install -P hudson,eap,pack200
>>>>> -Dtpc.version=4.50.0.Beta1-SNAPSHOT -DBUILD_NUMBER=450000
>>>>> -DJOB_NAME=devstudio.product_master
>>>>> -Dupdate.site.description="Development Milestone"
>>>>> -Djbosstools_site_stream=master -Dtpc.targetKind=unified
>>>>>
-DtargetRepositoryUrl=file:///home/nboldt/tru/targetplatforms/jbdevstudio/multiple/target/jbdevstudio-multiple.target.repo/
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>> --
>>>> Nick Boldt :: JBoss by Red Hat
>>>> Productization Lead :: JBoss Tools & Dev Studio
>>>>
http://nick.divbyzero.com
>>>> _______________________________________________
>>>> jbosstools-dev mailing list
>>>> jbosstools-dev(a)lists.jboss.org
>>>>
https://lists.jboss.org/mailman/listinfo/jbosstools-dev
>>>
>>>
>>> /max
>>>
http://about.me/maxandersen
>>
>> --
>> Nick Boldt :: JBoss by Red Hat
>> Productization Lead :: JBoss Tools & Dev Studio
>>
http://nick.divbyzero.com
>
>
> /max
>
http://about.me/maxandersen
--
Nick Boldt :: JBoss by Red Hat
Productization Lead :: JBoss Tools & Dev Studio
http://nick.divbyzero.com