While the deployer is down, you cannot deploy configuration bundle updates, such as apps, to the search head cluster members. Because all cluster members must use the same set of configurations, you must, with few exceptions, ensure that a member does not join or rejoin the cluster while the deployer is down. If no member joins the cluster and you have no immediate need to push any new configurations, the cluster can continue to operate without a deployer indefinitely. Otherwise, you must bring up a new deployer before a member joins or rejoins the cluster: Create a new deployer instance. Restore the contents of $SPLUNK_HOME/etc/shcluster to the new instance from backup. If necessary, update the conf_deploy_fetch_url values on all search head cluster members. Push the restored bundle contents to all members by running the splunk apply shcluster-bundle command. For details, see http://docs.splunk.com/Documentation/Splunk/6.2.2/DistSearch/PropagateSHCconfigurationchanges#Consequence_and_remediation_of_deployer_failure ... View more

You do not need to run the same version across clusters. However, within a 6.1 cluster, all instances (master, peers, search heads) must be running the same version. So, as long as you do not have a search head trying to access both clusters, you should be fine. Starting with 6.2, btw, the restrictions around intra-cluster version compatibility have been loosened to some degree. See: http://docs.splunk.com/Documentation/Splunk/6.2.2/Indexer/Systemrequirements#Splunk_Enterprise_version_compatibility ... View more

Do you have any data in the indexes? New indexes only show up in the Indexes tab of the Master Dashboard once they have data in them. See the "note" here: http://docs.splunk.com/Documentation/Splunk/6.2.1/Indexer/Howtomonitoracluster#Indexes_tab ... View more

You can implement indexer clustering even when replication is not required. There are some good reasons to do so, documented here: http://docs.splunk.com/Documentation/Splunk/6.2.0/Indexer/Clustersinscaledoutdeployments As you mention, this topology has additional benefits with the advent of search head clustering, particularly in regards to simplifying the connection to the search peers. ... View more