[
https://issues.apache.org/jira/browse/SLING-6543?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Konrad Windszus updated SLING-6543:
-----------------------------------
Description:
Right now the {{start}} goal will only start the server (in a dedicated
process) and then return when the control port returns that the server is
started.
That does not necessarily mean that all services from all bundles are already
fully started.
It should be possible to block until the server is fully started and some
services are up and running. Otherwise follow-up processes (like
{{maven-failsafe-plugin}}) will need to implement some wait approach.
One possibility is to issue repeated GET requests against the instance at a
dedicated URL until it reports a 200 (similar to what is done in
{{SlingTestBase.waitForServerReader(...)}}
(https://github.com/apache/sling/blob/trunk/testing/tools/src/main/java/org/apache/sling/testing/tools/sling/SlingTestBase.java#L263).
was:Right now the {{start}} goal will only start the server (in a dedicated
process) and then return. It should be possible to block until the server is
fully started. Otherwise follow-up processes (like {{maven-failsafe-plugin}})
will need to implement some wait approach and also logging of Maven is
intertwined with the logging of the starting server.
> slingstart-maven-plugin: Optionally make the start mojo blocking until the
> server is fully up and running
> ---------------------------------------------------------------------------------------------------------
>
> Key: SLING-6543
> URL: https://issues.apache.org/jira/browse/SLING-6543
> Project: Sling
> Issue Type: Bug
> Components: Tooling
> Affects Versions: Slingstart Maven Plugin 1.7.0
> Reporter: Konrad Windszus
>
> Right now the {{start}} goal will only start the server (in a dedicated
> process) and then return when the control port returns that the server is
> started.
> That does not necessarily mean that all services from all bundles are already
> fully started.
> It should be possible to block until the server is fully started and some
> services are up and running. Otherwise follow-up processes (like
> {{maven-failsafe-plugin}}) will need to implement some wait approach.
> One possibility is to issue repeated GET requests against the instance at a
> dedicated URL until it reports a 200 (similar to what is done in
> {{SlingTestBase.waitForServerReader(...)}}
> (https://github.com/apache/sling/blob/trunk/testing/tools/src/main/java/org/apache/sling/testing/tools/sling/SlingTestBase.java#L263).
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
