# Deploying for specific target environment/machine
**URL:** https://drone.discourse.group/t/deploying-for-specific-target-environment-machine/5243
**Category:** Drone Support
**Created:** [December 2, 2020, 9:44pm UTC](https://drone.discourse.group/t/deploying-for-specific-target-environment-machine/5243 "2020-12-02T21:44:52Z")
**Posts on this page:** 3
**Page:** 1
### Author:  [@misamae](https://drone.discourse.group/u/misamae)
#### Post date: [December 2, 2020, 9:44pm UTC](https://drone.discourse.group/t/deploying-for-specific-target-environment-machine/5243/1 "2020-12-02T21:44:52Z")
I have read in the documentation that a runner can be connected to the drone server via the `DRONE_RPC_HOST` and `DRONE_RPC_SECRET` environment variables.
However, I cannot find out how I can target each runner in my pipeline. In my scenario I would like to deploy merges to the main branch to on my production server and dev branch to my test sever.
---
### Author:  [@bradrydzewski](https://drone.discourse.group/u/bradrydzewski)
#### Post date: [December 2, 2020, 10:36pm UTC](https://drone.discourse.group/t/deploying-for-specific-target-environment-machine/5243/2 "2020-12-02T22:36:21Z")
Drone is not really meant to work this way. Drone was designed to run pipelines across a pool of available nodes, similar to travis or circle (where you don’t get to pick the node). If you want to deploy to a specific machine, you add a step to your pipeline that uses something like scp to upload artifacts to the machine, or ssh to login and execute commands on the machine.
---
### Author:  [@misamae](https://drone.discourse.group/u/misamae)
#### Post date: [December 3, 2020, 7:29am UTC](https://drone.discourse.group/t/deploying-for-specific-target-environment-machine/5243/3 "2020-12-03T07:29:47Z")
Thanks for the explanation. I think the following part can be added to runners documentation. It will help beginners like myself.
`Drone was designed to run pipelines across a pool of available nodes,`