Thanks for getting back to me.
We are moving very quickly at upgrading everything to Java 17 and as
security is our #1 concern, we have to be diligent and careful with
testing and updates. Since you have verified that what I found is a
valid failure (e.g. - as I suspected Java 17 is stricter than Java 11
and Java 8), we can ignore the failure (which I actually fixed in my
build by changing the hostname used in the failing test). Regarding the
script engine failure, that was easily fixed by enabling Nashorn support
in the GraalVM compiler.
A full regression test is taking some time, but so far Qpid seems to be
working fairly well as a native image.
Thanks,
PGA
On 10/10/2022 10:02 AM, Daniil Kirilyuk wrote:
Hello,
As far as I know, a switch to ignore the invalid hostname was added
with JIRA QPID-8535 (https://issues.apache.org/jira/browse/QPID-8535)
due to the errors arising when hostname has had an underscore
character. That seemed to be a workaround for a quite specific case.
The workaround mechanism was broken in Java 17 due to the changes in
SNI validation mechanism.
The unit test failing under Java 17 should be fixed with QPID-8587
(https://issues.apache.org/jira/browse/QPID-8587), PR -
https://github.com/apache/qpid-broker-j/pull/135.
We’re currently working on preparing the next Broker-J public build
which will include this change as well as some other fixes and
improvements (documented in latest Broker-J JIRAs), but it may take
some time till it will be available. We hope to manage it till the end
of the year.
Kind regards,
Daniil Kirilyuk
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@qpid.apache.org
For additional commands, e-mail: dev-h...@qpid.apache.org
