Re: [jakartaee-tck-dev] committers: The jakartaeetck-nightly-run-master

[Scott Marlow <smarlow@xxxxxxxxxx>]

On 7/1/20 4:42 PM, Steve Millidge (Payara) wrote:
> Looking at that run it is still validating resources so not sure the System property is set. You should hit this line  https://github.com/eclipse-ee4j/glassfish/blob/f5208511b0d1b4a24624b04895fa0b0f598848b3/appserver/deployment/dol/src/main/java/com/sun/enterprise/deployment/util/ResourceValidator.java#L101 but I can still see validateResource in the stack trace.

Thanks Steve, I need to spend some time setting up "podman" locally to 
run GlassFish (so I can verify changes like this locally).

https://github.com/eclipse-ee4j/jakartaee-tck/pull/352 is the new change 
to attempt this.

Scott

> -----Original Message-----
> From: Scott Marlow <smarlow@xxxxxxxxxx>
> Sent: 01 July 2020 21:32
> To: jakartaee-tck developer discussions <jakartaee-tck-dev@xxxxxxxxxxx>; Steve Millidge (Payara) <steve.millidge@xxxxxxxxxxx>
> Subject: Re: [jakartaee-tck-dev] committers: The jakartaeetck-nightly-run-master job will now prompt for the tests to run
>
>
>
> On 7/1/20 10:59 AM, Scott Marlow wrote:
> > On 7/1/20 6:54 AM, Steve Millidge (Payara) wrote:
> > > I’m not sure. I recognise the stack trace coming out of the
> > > deployment validator. In Payara, a few years back, we fixed bugs in
> > > this area whereby the server is attempting to validate JNDI resources
> > > during deployment that are not there yet as they haven’t been created
> > > yet in the deployment. I’m wondering whether the switch to
> > > auto-deployment is hitting those bugs.
> >
> > I created a gist of the failure at
> > https://gist.github.com/scottmarlow/6272a32cef23585696a17fce642b21f3
> > for future reference.
> >
> > I did a diff of changes we made for EE 9 Platform TCK and do not see
> > any references to "deployment.resource.validation".
> >
> > As you suggest, we can try working around what might be a deployment
> > ordering issue, caused by the switch to deploying via file system
> > (which really should work).  Workaround to try is
> > -Ddeployment.resource.validation=false.
>
> https://ci.eclipse.org/jakartaee-tck/job/jakartaeetck-nightly-run-master/54/
> is running with the change to use -Ddeployment.resource.validation=false
> for tests that use Jakarta Mail tests.
>
> > Scott
> >
> > The system property switches off this resource
> > > validation which isn’t necessary.
> >
> >
> >
> > > Just a hunch.
> > >
> > > Steve
> > >
> > > *From:*jakartaee-tck-dev-bounces@xxxxxxxxxxx
> > > <jakartaee-tck-dev-bounces@xxxxxxxxxxx> *On Behalf Of *Scott Marlow
> > > *Sent:* 01 July 2020 02:56
> > > *To:* jakartaee-tck developer discussions
> > > <jakartaee-tck-dev@xxxxxxxxxxx>
> > > *Subject:* Re: [jakartaee-tck-dev] committers: The
> > > jakartaeetck-nightly-run-master job will now prompt for the tests to
> > > run
> > >
> > > On Tue, Jun 30, 2020 at 4:32 PM Steve Millidge (Payara)
> > > <steve.millidge@xxxxxxxxxxx <mailto:steve.millidge@xxxxxxxxxxx>> wrote:
> > >
> > >      Some of those mail session failures MAY be fixed by adding
> > >      -Ddeployment.resource.validation=false To the GlassFish startup.
> > >
> > > Any idea what the underlying cause is?
> > >
> > > Could it be a problem with a dtd reference in the test, like
> > >
> > > https://github.com/eclipse-ee4j/jakartaee-tck/blob/master/src/com/sun
> > > /ts/tests/javamail/ee/getMessageContent/getMessageContent_appclient_v
> > > ehicle_client.jar.sun-application-client.xml#L2
> > >
> > >
> > > ?
> > >
> > >
> > > _______________________________________________
> > > jakartaee-tck-dev mailing list
> > > jakartaee-tck-dev@xxxxxxxxxxx
> > > To unsubscribe from this list, visit
> > > https://www.eclipse.org/mailman/listinfo/jakartaee-tck-dev