4 months ago
I want clarification about your advertised resource limits.
You say that for $20/month I get 32 CPU cores and 32 GB RAM for every deployed service.
I tested this by deploying a single service with the displayed limits: 32 CPU / 32 GB RAM. Then I tried running multiple browser instances inside it.
If the service really had 32 cores and 32 GB RAM, it should easily run around a hundred browser instances or even more. But in practice, it can’t even handle 5. After that, the service starts crashing, and new browser processes fail to spawn.
Your own metrics also show strange numbers: only about 2 CPU cores in use and around 6–8 GB RAM used before everything breaks. It doesn’t behave like a machine with the resources you claim.
So I need an explanation:
Why does a service that is advertised as 32 CPU / 32 GB RAM perform like a much smaller machine? Is the displayed resource allocation not real, or is there some kind of throttling or hidden limits?
Please clarify what is actually happening.
5 Replies
4 months ago
Hey there! We've found the following might help you get unblocked faster:
If you find the answer from one of these, please let us know by solving the thread!
4 months ago
i think we are facing the same issue, take a look at usage metrics, its probably not using alot of CPU and memory, at least on my side it isnt
4 months ago
Hello!
I can completely understand the frustration on that front. That being said, the numbers we advertise are the numbers we provide! The Railway pro plan has 32g of ram and 32vCPU available for each of your services.
I suspect you might be running into a PID limit however I would want to be sure. Do you have a time range and timezone of when your service was running into issues? Would like to investigate logs.
Status changed to Awaiting User Response Railway • 4 months ago
4 months ago
It was happening around November 17 00:50 in the NstBrowser service. You can see a lot of such logs:
[31mERRO[2025-11-16 21:50:56.279] browser start error: fork/exec /root/.nst-agent/download/kernels/nstchrome/nstchrome-140-202509241200/nstchrome: resource temporarily unavailable[0m
Status changed to Awaiting Railway Response Railway • 4 months ago
4 months ago
My suspicion of PID limit looks to be the culprit.
Are you able to monitor active PID count and see what it sits at? Right now we have a limit of 1k.
Status changed to Awaiting User Response Railway • 4 months ago
4 months ago
This thread has been marked as solved automatically due to a lack of recent activity. Please re-open this thread or create a new one if you require further assistance. Thank you!
Status changed to Solved Railway • 4 months ago