I haven't come across this specific issue but might I suggest you enable traces for the CDR Repository manager service and then see if the root cause becomes any clearer. You can enable tracing throught Unified Serviceability. Go to Trace > Configuration. Select your publishe node, then select the CDR Services Service Group, and finally select the CDR Repository Manager service.
Set your trace level to detail. Then I would restart the service (or attempt to start it) under Tools > Control Center - Network Services. This should generate some content in the trace file, which you can then view and/or download via RTMT.
Since you are doing a restore, one of the first things that comes to mind for me is related to the ipsec security certificates. I know that DRS and some other cluster functions can be affected when doing a DRS restore on a new cluster, even though the certificates from the DRS backup (previous instance of your UCM cluster) are supposed to be restored. I have resolved the DRS issues I have seen by regenerating the ipsec.cer self-signed cert and then restarting affected services. I am not sure if this is relevant for CDR Repository Manager. It is just the first thing that comes to mind.
Anyway, look at the trace files (after you elevate the trace level to detailed). You may also want to look at the application syslog (via RTMT). Somewhere in there should be some clues as to what is going on. If you have a hard time interpreting the logs/traces then post back and someone in the forum should be able to assist.
HTH
-Bill
(b) http://ucguerrilla.com
(t) @ucguerrilla
Please remember to rate helpful responses and identify helpful or correct answers.
