Thanks John and Romain,
Will work on the new FailsafeExecutionManagerProvider.
Cheers
Bruno Baptista
http://twitter.com/brunobat_
On 03/10/2018 18:21, John D. Ament wrote:
Hi Bruno
Thanks for the PR!
I think my intention for what's provided in Safeguard is that we have
an overideable per container integration that allows you to look up
the executor. So rather than having boolean logic, you use a new
implementation of FailsafeExecutionManagerProvider (perhaps as an
alternative). This way the lookup can be done based on how your
platform is developed.
Thanks,
John
On Wed, Oct 3, 2018 at 5:53 AM Romain Manni-Bucau
<[email protected] <mailto:[email protected]>> wrote:
yes, this is why I mentionned to make the pool configurable to
make it work in both environment and in multiple apps with
different pool.
Out of my head I thought about making it injectable instead of
trying all possible strategies/relying on a system properties but
I just realized that we already support SE and EE with managed
pool, just make a @Specializes of FailsafeExecutionManagerProvider
producer.
It sounds to me more flexible and easier to understand.
wdyt?
Romain Manni-Bucau
@rmannibucau <https://twitter.com/rmannibucau> | Blog
<https://rmannibucau.metawerx.net/> | Old Blog
<http://rmannibucau.wordpress.com> | Github
<https://github.com/rmannibucau> | LinkedIn
<https://www.linkedin.com/in/rmannibucau> | Book
<https://www.packtpub.com/application-development/java-ee-8-high-performance>
Le mer. 3 oct. 2018 à 11:39, Bruno Baptista <[email protected]
<mailto:[email protected]>> a écrit :
Hi Romain,
I've updated the PR to get the resource location from a property.
In relation to the try/catch... I'm assuming that the library
is supposed to work with both SE and EE environments, if we
don't catch the exception this will never work on SE. In any
case, if an error exists, it will be printed and can be found.
Cheers
Bruno Baptista
http://twitter.com/brunobat_
On 03/10/2018 10:10, Romain Manni-Bucau wrote:
Hi Bruno,
Technically this pool does not "have to be" managed ;).
That said this is a good feature. Can you make the pool
configurable instead of hardcoding the default pool which is
never used except in tests? Will also avoid to catch and
silently ignore the error (can be an issue in servers).
Romain Manni-Bucau
@rmannibucau <https://twitter.com/rmannibucau> | Blog
<https://rmannibucau.metawerx.net/> | Old Blog
<http://rmannibucau.wordpress.com> | Github
<https://github.com/rmannibucau> | LinkedIn
<https://www.linkedin.com/in/rmannibucau> | Book
<https://www.packtpub.com/application-development/java-ee-8-high-performance>
Le mer. 3 oct. 2018 à 10:58, Bruno Baptista
<[email protected] <mailto:[email protected]>> a écrit :
Hi folks,
Safegard creates a java se connection pool to handle the
bulckhead and
async operations. When deployed in a container, that pool
has to be
managed.
I created a PR that allows to retrieve that managed pool,
if available:
https://github.com/apache/geronimo-safeguard/pull/2
Can someone please take a look at it?
Regards
--
Bruno Baptista
http://twitter.com/brunobat_
