# Marvel agent not sending to Marvel monitoring cluster **URL:** \ **Category:** Elasticsearch\ **Tags:** elastic-stack-monitoring\ **Created:** [January 12, 2016, 5:35pm UTC](https://discuss.elastic.co/t/marvel-agent-not-sending-to-marvel-monitoring-cluster/39016 "2016-01-12T17:35:27Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![chapeaurouge](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/chapeaurouge/32/7104_2.png) [@chapeaurouge](https://discuss.elastic.co/u/chapeaurouge)\ **Post date:** [January 12, 2016, 5:35pm UTC](https://discuss.elastic.co/t/marvel-agent-not-sending-to-marvel-monitoring-cluster/39016/1 "2016-01-12T17:35:28Z")
Hi all, First time around trying to setup an external Marvel cluster for monitoring several other clusters. I have setup a 3-nodes Marvel cluster based on the latest version, ES 2.1.1, Kibana 4.3.1, Marvel 2.1.1. The first production cluster I want to monitor and that needs to ship its data is running on ES 1.7.3. Others are running in 1.7.x for the moment. My Marvel cluster is running OK, and is currently indexing its own data (just to have something to show...) I have configured each node of my prod cluster to export via http, with old fashion, such as: marvel.agent.export.es.hosts: ["[http://node1:9200](http://node1:9200)", "[http://node2:9200](http://node2:9200)"] and also the "new" way: marvel.agents.exporters: id1: type: http host: ["[http://node1:9200](http://node1:9200)", "[http://node2:9200](http://node2:9200)"] However, nothing shows up in my Kibana Marvel app. Also, nothing in the logs of my prod cluster nodes related to something happening with regards to marvel.agent.exporter. It is like my configuration has no effect. Could you guys advise? I would think that my Marvel cluster components can be of a higher version than the nodes I want to monitor.. am I wrong? Thanks a lot for your time and help. Cheers, fred ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [January 12, 2016, 9:50pm UTC](https://discuss.elastic.co/t/marvel-agent-not-sending-to-marvel-monitoring-cluster/39016/2 "2016-01-12T21:50:20Z")
You can't do cross versions like that unfortunately. So your 1.7.X cluster needs the same on the other end. ---
**Author:** ![chapeaurouge](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/chapeaurouge/32/7104_2.png) [@chapeaurouge](https://discuss.elastic.co/u/chapeaurouge)\ **Post date:** [January 13, 2016, 9:35am UTC](https://discuss.elastic.co/t/marvel-agent-not-sending-to-marvel-monitoring-cluster/39016/3 "2016-01-13T09:35:43Z")
OK, thanks for the quick reply. Given what I see now, should I expect to build a monitoring cluster for each production cluster I have? Thanks, Best. fred ---
**Author:** ![warkolm](https://sea2.discourse-cdn.com/elastic/user_avatar/discuss.elastic.co/warkolm/32/39224_2.png) [@warkolm](https://discuss.elastic.co/u/warkolm)\ **Post date:** [January 13, 2016, 10:23am UTC](https://discuss.elastic.co/t/marvel-agent-not-sending-to-marvel-monitoring-cluster/39016/4 "2016-01-13T10:23:42Z")
Unless you are a subscription cluster, yes. Subscription customers get access to multi-cluster Marvel (ie one Marvel cluster, multiple monitored ones). ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 6, 2017, 1:45pm UTC](https://discuss.elastic.co/t/marvel-agent-not-sending-to-marvel-monitoring-cluster/39016/5 "2017-07-06T13:45:43Z")