Locate the Nagios persistent volume claim-name from the Nagios pod:

oc describe pod <nagios-pod-name> | grep "ClaimName:"
claimName: <claim-name>

Determine the volume-name using the claim-name from step 1:

oc describe pvc <claim-name> | grep "Volume:"
Volume: <volume-name>

Using the volume-name, run the following command:

oc describe pv <volume-name>

The output contains a Source section, similar to:

Source:
    Type:      	NFS (an NFS mount that lasts the lifetime of a pod)
    Server:    	nfs-server.local
    Path:      	/path/to/nfs/directory
    ReadOnly:  	false

This information describes where the data for the Nagios persistent volume is located.

Stop the Nagios pod by editing the Nagios deployment config and set replicas to 0:

oc edit dc nagios

Change replicas to 0:

spec:
  replicas: 0

Once the restore is complete, start the Nagios pod by editing the Nagios deployment config and set replicas to 1:

oc edit dc nagios

Change replicas back to 1:

spec:
  replicas: 1

Locate the primary Mongodb node. You must restore using the primary member of the Mongodb replicaset, you can find this by logging into any of the Mongodb nodes via remote shell:

oc rsh <mongodb-pod-name>

Connect to the mongo shell in this pod:

mongo admin -u admin -p ${MONGODB_ADMIN_PASSWORD}

Check the replicaset status:

rs.status()

The node with the "stateStr" key set to PRIMARY is the node to use in the following restoration procedure.

Upload MongoDB data to the primary pod using the oc rsync command:

oc rsync /path/to/backup/directory/ <primary-mongodb-pod>:/tmp/backup

Import data into the Mongodb replicaset using the 'mongorestore' tool from inside the mongodb pod. Log into the pod:

oc rsh <mongodb-pod-name>

Restore the data:

mongorestore -u admin -p ${MONGODB_ADMIN_PASSWORD} --gzip --archive=/tmp/backup/mongodb-backup.gz

For more information on this tool, see the mongorestore documentation.