Refresh Snapshot Process Doesn't Start Because of Concurrent Manager Unavailability

When launching Collections, the Planning Data Pull fails with the following error: "Refresh Snapshot process could not get started after waiting for X minutes because of Concurrent Manager Unavailability". This can happen in a centralized instance when dblink information has been populated in msc_apps_instances inadvertently. When dblink information is populated, the Planning Data Pull tries to launch the refresh collection snapshot using the dblink information available, but it fails to find a connected instance, and finally times out.
To resolve, run the following sql:

SELECT instance_id,instance_code, 
nvl(a2m_dblink,'Null'),nvl(m2a_dblink,'Null') 
FROM msc_apps_instances ;

If the above sql returns a value for a2mdbling and m2adblink other than the value 'Null', then the setup is wrong, and you need to do the following:

1. Delete the records from the table MRP_AP_APPS_INSTANCES_ALL as follows:
   

   DELETE
   FROM mrp_ap_apps_instances_all
   WHERE instance_code = '&instance_code'; 

   Commit;

2. Go into the Advanced Supply Chain Planning Administrator responsibility.
   
3. Navigate to Admin > Instances.
   
4. Remove the dblink information. This will change the record in msc_apps_instances, mrp_ap_apps_instances_all and mrp_ap_apps_instance.
   
5. Confirm whether the dblink information has correctly populated as Null in the msc_apps_instances using the following sql:
   

   SELECT instance_id,instance_code,
   nvl (a2m_dblink,'Null'),nvl (m2a_dblink,'Null')
   FROM msc_apps_instances;