Hello, My use case : Context : On azure, datas from several applications are pushed in a Azure EventHub I need to separate the datas from one application, and put this datas into a new index on Splunk On Azure, all the resources of this app are in one Ressource Group : TheAppResourceGroupName I used a Heavy Forwarder, and this are my configs : props.conf : [source::eventhub://EVENTHUBAZURE.servicebus.windows.net/app-logs;] TRANSFORMS-route =  routeToNewIndex, discard_original, transforms.conf [routeToNewIndex] REGEX = TheAppResourceGroupName DEST_KEY = _MetaData:Index FORMAT = NewIndex [discard_original] REGEX = TheAppResourceGroupName DEST_KEY = queue FORMAT = nullQueue This config will delete the datas, yes, but in the NewIndex, and not in the original Index, after the routing. I didn't find an answer witch fit with my needs on the commu and the docs, but maybe someone has to face a similar need . Thanks a lot for the help! Nico ... View more

Hi all After temptative for troubleshooting my issue alone, I will try my luck here. Purpose : clone one sourcetype to store the logs into a local indexer, and in a distant one I use one heavy forwarder to receive the logs, store the logs in a indexer, and same heavy forwarder will clone the sourcetype to forward the cloned one into a distant heavy forward, that I don't managed. Here is my config : [inputs.conf] [udp://22210] index = my_logs_indexer sourcetype = log_sourcetype disabled = false This works pretty well, and all logs are stored into my indexer Now will come the cloning part :  [props.conf] [log_sourcetype] TRANSFORMS-log_sourcetype-clone = log_sourcetype-clone [transforms.conf] [log_sourcetype-clone] REGEX = . DEST_KEY = _TCP_ROUTING FORMAT = distant_HF_formylogs [outputs.conf] => for cloned logs [tcpout:distant_HF_formylogs] server = ip_of_distant_HF:port sendCookedData = false This configuration is used for another use case, as sometimes I have had to anonymize some logs. However, for this particular use case, when I activate the cloning part, it stops the complete log flow for this use case, even on the local indexers. I didn't quite understand why, because I don't see the difference with my other use case, apart from the fact that the logs are UDP logs and not TCP. Am I missing something? Thanks a lot for your help ... View more