Full Remote Mode
With full remote mode, you can work on a macOS, Linux, or Windows desktop targeting a remote Linux machine connected via SSH. You can choose any Linux-based target, including embedded systems on single-board computers like Raspberry Pi. Also, your program can be launched on a cloud platform or, for example, inside a Docker container.
For remote development, the CLion instance runs locally, and your source files are also placed on the local client, with automatic synchronization to the remote host. On the remote host side, CLion performs compilation and build using host compilers and CMake, uses host GDB for debug, and runs the application on the remote target.
When working on a Windows client, keep in mind the following:
Due to the IntelliJ platform issue, you need to set the property value
idea.case.sensitive.fs=truein the idea.properties file (to access the property file, select on the main menu), then restart CLion with cache reset ( ).
For files synchronization on Windows, CLion relies on its own Remote Host Access and compression on the host side using the tar utility. This mechanism causes the synchronization to perform slower than rsync on macOS and Linux.
Create a toolchain with remote credentials
Go toand click to create a new toolchain.
Select Remote Host from the drop-down menu and click in the Credentials field. In the dialog that opens, provide the credentials for accessing you remote machine:
After establishing the connection, CLion attempts to detect the tools in default remote locations /usr/bin/cmake and /usr/bin/gdb (or using the full paths, if you have provided manually). When the checks finish successfully, the toolchain is ready for use:
You can make the newly created toolchain the default one (for this, move it to the top of the toolchains list by clicking ). When set as default, the remote toolchain is used for all the projects you create and open in CLion.
Note that if you set the remote toolchain as default, the default CMake profile will connect to it automatically, so you do not need to configure a separate CMake profile for it.
Create the corresponding CMake profile
Go to CMake profile, and connect it to your remote toolchain using the Toolchain field:, click to create a new
Check and adjust the deployment configuration
When you create a connection entity for the remote toolchain, CLion puts it in the list of server access configurations in :
CLion automatically configures the paths for your project code synchronization. Use the Mappings tab to change the default mappings (for example, to set a particular remote directory for the copied sources instead of the default tmp folder:
You can monitor the synchronization process in the File Transfer tool window ( ):
Resync header search paths
To resolve your code correctly, CLion synchronizes header search paths with all the content from the remote machine to the local client. For example, even though standard library headers are taken from the target, you can navigate to them as if you were working locally in the CLion editor.
However, header search paths synchronization can be time-consuming, so CLion performs it automatically only upon the initial file transfer. After that, it is not triggered by CMake reloads. So every time you switch the compiler or make changes in your project dependencies, make sure to update header search paths manually by calling.
You can also switch to automatic synchronization: set the clion.remote.resync.system.cache key in the Registry (go toor press Ctrl+Shift+A, type Registry, and search for the key by name).
Build, run, debug
Now that you have a remote toolchain and the corresponding CMake profile configured, you can build, run, and debug your application and tests in the completely remote way by selecting the proper CMake profile in the Run/Debug configuration switcher:
The demo below shows how the application output changes depending on the OS that it runs on. Having macOS as a local system, we connect remotely to the Ubuntu target and check the OS name. In this example, code highlighting depends on the OS identifier. That is why when we switch the CMake profile, and the resolve context changes accordingly, or when we switch the resolve context manually in the right-bottom corner of the status bar, we can see the corresponding code branch being highlighted:
Enable IPv6 support
Run *.vmoptions file that opens, delete thefrom the main menu. In the
-Djava.net.preferIPv4Stack=trueline and add the following lines:-Djava.net.preferIPv4Stack=false -Djava.net.preferIPv6Stack=true -Djava.net.preferIPv6Addresses=true
In order to use hostnames instead of raw addresses, open C:\Windows\System32\Drivers\etc\hosts as Administrator on Windows or /etc/hosts as superuser on macOS/Linux and map the required addresses to the corresponding hostnames.
Each address should be placed on a separate line, followed by at least one whitespace and a list of whitespace-separated hostnames, for example:f381::171d:c61c:c7f3:3a56 my.dev.host1 my.dev.host2 f381::171d:c61c:c7f3:3a26%en0 my.dev.host3
On macOS, you also need to specify the list of the banned network interfaces:
Go toand type Registry.
Find the deployment.macOs.bannedInterfaces key and set a comma-separated list of the interfaces to be banned, for example