# Multi instance configuration **URL:** \ **Category:** Elasticsearch\ **Created:** [June 2, 2016, 1:36pm UTC](https://discuss.elastic.co/t/multi-instance-configuration/51702 "2016-06-02T13:36:43Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![gphadke](https://avatars.discourse-cdn.com/v4/letter/g/ecc23a/32.png) [@gphadke](https://discuss.elastic.co/u/gphadke)\ **Post date:** [June 2, 2016, 1:36pm UTC](https://discuss.elastic.co/t/multi-instance-configuration/51702/1 "2016-06-02T13:36:43Z")
Hello I am trying to setup multiple instances running on same physical box. Among the multiple instances I would like to set 1 as master and others as data only. I was planning to have the data nodes 1,2 listen on ports 9201 and 9202 respectively and always. I want to give the connections settings of data nodes only to the es clients. I have the below config Master http.port: 9200 transport.tcp.port: 9300 network.host: ["192.168.100.102", _local_] discovery.zen.ping.timeout: 3s discovery.zen.ping.multicast.enabled: true discovery.zen.ping.unicast.hosts: ["192.168.100.102"] cloud-2: node.master: false node.data: true node.master: true node.data: false http.port: 9201 transport.tcp.port: 9301 network.host: ["192.168.100.102", _local_] discovery.zen.ping.timeout: 3s discovery.zen.ping.multicast.enabled: true discovery.zen.ping.unicast.hosts: ["192.168.100.102"] cloud-3: node.master: false node.data: true http.port: 9202 transport.tcp.port: 9302 network.host: ["192.168.100.102", _local_] discovery.zen.ping.timeout: 3s discovery.zen.ping.multicast.enabled: true discovery.zen.ping.unicast.hosts: ["192.168.100.102"] What I see is with these settings for the data only nodes (cloud-2 and cloud-3), It fails to discover the master running on 9300. The cloud-3 instance seems to be pinging its own self. Same with cloud-2 If I comment out http.port and transport.tcp.port, the discovery seems to work, but the instances select first available ports in the range. How can I set pre-defined ports for http and transport through yml as I am configuring these as connection variables in other ES clients. [discovery.zen.ping.unicast] [cloud-3] [1] sending to {cloud-3}{VJtxMqsxSYCCvTRDCjHxLQ}{192.168.100.102}{192.168.100.102:9302}{rack\_id=cloud, master=false} [2016-06-02 13:26:09,514][TRACE][discovery.zen.ping.unicast] [cloud-3] [1] received response from {cloud-3}{VJtxMqsxSYCCvTRDCjHxLQ}{192.168.100.102}{192.168.100.102:9302}{rack\_id=cloud, master=false}: [ping\_response{node [{cloud-3}{VJtxMqsxSYCCvTRDCjHxLQ}{192.168.100.102}{192.168.100.102:9302}{rack\_id=cloud, master=false}], id[1], master [null], hasJoinedOnce [false], cluster\_name[vagrant\_low\_profile]}, ping\_response{node [{cloud-3}{VJtxMqsxSYCCvTRDCjHxLQ}{192.168.100.102}{192.168.100.102:9302}{rack\_id=cloud, master=false}], id[2], master [null], hasJoinedOnce [false], cluster\_name[vagrant\_low\_profile]}] I am using elasticsearch 2.3.1 ---
**Author:** ![system](https://us1.discourse-cdn.com/elastic/original/3X/1/a/1ac57faf039f6b580b3f104ef42a2a89e41014de.png) [@system](https://discuss.elastic.co/u/system)\ **Post date:** [July 5, 2017, 10:46pm UTC](https://discuss.elastic.co/t/multi-instance-configuration/51702/2 "2017-07-05T22:46:52Z")