[
https://issues.apache.org/jira/browse/CXF-7490?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16146952#comment-16146952
]
Max Schlüssel commented on CXF-7490:
------------------------------------
Also, when deploying that bundle, only in Linux the following message appears
in the log:
{code:log}
2017-08-30 11:25:52,754 | WARN | aba/karaf/deploy |
NamespaceHandlerRegistryImpl | 12 -
org.apache.aries.blueprint.core - 1.7.1 | NamespaceHandler
org.apache.cxf.jaxrs.blueprint.JAXRSBPNamespaceHandler is behaving badly and
should be fixed
2017-08-30 11:25:52,761 | WARN | aba/karaf/deploy |
NamespaceHandlerRegistryImpl | 12 - org.apache.aries.blueprint.core - 1.7.1
| Unable to find namespace handler for http://cxf.apache.org/configuration/beans
{code}
> JAX-RS REST server causes long bundle starting time in OSGi blueprint
> ---------------------------------------------------------------------
>
> Key: CXF-7490
> URL: https://issues.apache.org/jira/browse/CXF-7490
> Project: CXF
> Issue Type: Bug
> Components: JAX-RS, OSGi
> Affects Versions: 3.1.8
> Environment: Camel 2.18.1, Aries Blueprint 1.7.1, Karaf 4.0.9, Debian
> 3.16.39-1
> Reporter: Max Schlüssel
> Priority: Minor
> Attachments: my-rest-test.zip
>
>
> I've encountered a weird problem when creating a REST server in blueprint.
> The bundle has nothing but a simple server that contains one service bean,
> configured like this:
> {code:xml}
> <jaxrs:server address="/test" id="testRestService">
> <jaxrs:serviceBeans>
> <ref component-id="defaultTestRestService" />
> </jaxrs:serviceBeans>
> </jaxrs:server>
> <bean id="defaultTestRestService"
> class="de.maxdev.resttest.DefaultTestRestService" />
> {code}
> When I deploy this bundle on Karaf in a Linux environment, the bundle remains
> in the "Starting" state for about 120 seconds. On Windows however, it starts
> in about 3 seconds. When not taking care of bundle deployment order, this
> will block all the other bundles, too.
> Is this a bug or a misconfiguration? Maybe this is caused some lower-level
> socket issue?
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
