Can Rider run multiple instances?
When I open multiple solutions, I notice that Rider is single process.
I want to open muiltiple Rider instances, one for each solution. How can I do?
OS is Win10, Rider revision is 2023.3.4
Please sign in to leave a comment.
Hello,
When opening a new solution, another instance of Rider.Backend process will be launched. Regrettably, it is not possible to spawn a new frontend process (rider64.exe) for every solution. This restriction is derived from the IntelliJ platform, upon which Rider is based. As far as I'm aware, there are no plans to change it in the immediate future. Here is the relevant request in our issue tracker: IDEABKL-5934, feel free to upvote it.
Thanks Ivan!
open command prompt go into solution or folder, type “code .” go to other folder/solutions type “code .” as many editors and solutions open as you need.
Its not 1990, projects have Unity front ends, Unreal back ends and nodejs server code between them and you MUST have 3 different copies of your editor open in each of those projects AT THE SAME TIME.
If your editor cannot do this, get one which can, seems like rider may be good for little config file editing tasks, but not serious professional work.
Thank you for contacting Rider support.
That should work. You can open several Rider windows with different projects using the cli ("rider /path/to/the.sln") or using the "File | Open" action.
If that doesn't help, would you mind elaborating on the issue?
Have a nice day!
Hello Dmitry Kazantsev,
Why we cannot open new window directly via simply click to rider icon on desktop? Each time click file and open etc.
Hello,
Unfortunately, it isn't possible due to limitation of IDE architecture. No multiple instances of the same Rider version (or any other IJ-based IDE) can be launched in parallel via launcher.
Meantime, there are multiple ways of opening another Rider window above the File | Open. I listed these options here: SUPPORT-A-726 How to open several windows of JetBrains Rider
Hope that helps!
This is a major limitation in Rider IDE at the moment especially using AI. A nice workflow is too have “smaller” project that can be iterated upon using agentic AI per IDE instance. This works well for me in Visual Studio but unfortunately because of the lack of multiple instances in Rider IDE I find myself all the time switching back to visual studio. In AI iterations it is crucial to be able to iterate in different projects at the same time as AI agentic processes takes time to complete the desired actions therefor switching between instances is the only way to stay productive without “waiting” for AI all the time.
You guys need to seriously pay attention to this, or you are going to fall of the bus in AI development as current IDE restrictions (single instance) is severely limiting.
Hello Johan,
Thank you for your feedback. Do you experience any issues with agents when running several Rider windows for different projects?
You might want to try running multiple standalone Rider instances using this instruction: SUPPORT-A-2618 How to launch same Rider version multiple times as standalone processes? Do the same problems you are experiencing occur in such a setup?
Have a nice day!
Thanks Dmitry,
Honest, no I have no tried that as it easier just to use VS Code or VS for other IDE instances. I'm not desperate to force everything through Rider IDE atm as I have other options. Just mentioning this so that Rider is aware this issue causes a lot of friction when using AI agents specifically.
In my current multi monitor setup I like to use different Windows 11 “desktops/workspaces” Therefor in workspace 2 I use Rider IDE, in workspace 3 I will use say VS on a different project , workspace 4 I will use something else. Therefor when busy with one project and the ai agent is “busy/thinking” then I flick through to the other desktops and continue work there so that I minimize waiting for ai agents all the time with the great advantage that I no longer have to resize / minimize and refocus IDE's. Also this works well with AI agents as I have much smaller context window solving specific issues without having one large project. Works great !! This is also an area where Rider IDE is severely limiting, therefor my feeling that you guys will have to solve this as I think it's critical in “instance driven ai agent solutions”
Before AI Agents I was very happy with running a single instance of the IDE, post AI agents this scenario rapidly changed for me as I want to develop in a small context window per ai agent therefor multiple instances of IDE's.
Thank you for the detailed answer. I see your point. So, is your main concern that you can't launch another Rider instance via its launcher (from another desktop)? It is possible to open another Rider window from the running instance and move it to the necessary desktop.
Or you could make a per-desktop Rider launcher SUPPORT-A-2618, so that Rider will do run in different instances on each desktop.
Or is there some other issue with using agents inside a single Rider instance with several windows?
That is correct. You can't create different Rider IDE instances on the different desktops with ease without jumping through many hoops. I understand you have a workaround for instances but that is not a very user-friendly way in the current AI environment and comparing that to “cd" into cli path and ”code ." and you are good to go with another instance in VS Code for example.
I think the point I'm trying to make is it would be nice to “easily” create multiple instances of Rider IDE's wherever you want it in Windows.(and also in windows (virtual) desktops)
I see your point. Thank you for the feedback!
Greetings, all,
I would also point out that every other major IDE (VSCode, Visual Studio) for .Net supports true multi-instancing. I'm not a Java dev, but I would suspect that is true for Java IDEs other than IJ. Come in to the 21st century, folks. We have cookies!
Thanks,
Morgan Haldane
I would like to chime in with my own experience as to why Rider should be multi-instance by default.
For one, simply opening Rider using different launchers means I lose _all_ of my settings on the new instance (no solution history, all my plugins are gone, all my settings reset) so doing this at all is quite annoying. For another, it means it's on me to figure out which shortcut I used to open the current instance and that's just frustrating when I'm wrong and two solutions share the same settings.
Why is it frustrating to use 2+ solutions with the same profile? Well, say I want to use AI Assistant in one solution and not the other. When I open the pane in one and engage with it, when I give focus to the other window, the AI Assistant opens though I don't want it there. When I minimize it in #2, it minimizes in #1 once I click in that solution again (though I'm actively using it there). That's frustrating. Or if I'm interacting with dotCover in one solution, but not the other - I have to open it in both or else when I click between solutions, it'll close it though I'm still using it.
This is infuriating to experience in a modern IDE. Neither VS Code nor Visual Studio have this problem and it's a real thorn in my side when trying to do different development in different solutions at the same time when using Rider.
Whiteboy Thanks for your feedback. The problem you encountered could be mitigated by the following way:
Go to Window menu | Layout settings | Choose “Use unique Layout for Every Project”:
Please let me know if this does not work for you.
Thanks,
Tom
Well, that's not my username, but I I have found that setting in my instance and it has dramatically improved my experience already. Thank you for calling it out!