If the Fabric Maven Agent tries to resolve an artefact that is not available on any internal or external Maven repositories, it retries 9 times with exponentially growing delays between each retry. If the Maven artefact cannot be resolved in the first attempt, there should not really be a need to retry. Due to these retries and the exponentially growing delay (noticed up to 60 seconds) we delay the provisioning error up to 2 minutes for a single artefact and more for multiple artefacts.

The relevant code seems to be at
https://github.com/jboss-fuse/fabric8/blob/1.2.0.redhat-6-3-x/fabric/fabric-agent/src/main/java/io/fabric8/agent/download/impl/AbstractRetryableDownloadTask.java#L43-L50

Under debug logging I see:

 Error downloading mvn:com.mycompany/my-artefact/1.0.0: Error resolving artifact com.mycompany/my-artefact/1.0.0: Could not find artifact com.mycompany/my-artefact:jar:1.0.0 in repo_552881743 (http://290710de0b10:8181/maven/download/). Retrying in approx 48220 ms.

I suggest for Maven artefacts we don't need this retry logic.