19 hours ago
Description:
I upgraded my workspace plan to Hobby, which should provide 5GB of memory per replica. However, my Enarm-bot service is still limited to approximately 1GB (0.99999744 GB actual limit), causing the service to crash due to memory exhaustion.
Service Details:
Service ID: 7b999bc3-5469-4717-a1e5-3b5d6f43e000
Service Name: Enarm-bot
Project ID: bf448826-23e5-4274-81d9-6103ea55379b
Environment: production
Current Metrics:
Memory Usage: ~1GB (99.9% of limit)
Memory Limit: ~1GB (should be 5GB for Hobby plan)
Service Status: Crashed (1 of 1 replicas)
I have restarted the service and redeployed, but the memory limit did not update. The plan appears to be set correctly in the dashboard, but the backend is not applying the increased memory allocation.
Please verify that the Hobby plan upgrade was processed correctly and apply the correct 5GB memory limit to this service.
7 Replies
19 hours ago
This thread has been opened as a bounty so the community can help solve it.
Status changed to Open Railway • about 19 hours ago
19 hours ago
The backend didn't automatically apply the new 5GB limit to your Enarm-bot service. In settings-> resources tab, manually adjust the memory allocation in the settings for that specific service to ensure the correct amount is assigned.
darseen
Try to delete your service, and create a new one.
19 hours ago
That is unnecessary for configurations and variables. Also manual adjustment of memory (replica limits) in service settings->deploy->replica limits is advised in railway official docs.
19 hours ago
Memory limit for hobby plan is 8GB, 5GB is limit for volume
If you're talking about service memory limit, you can go to service settings and increase the replica limit on that page
But if you're talking about volume (which has 5GB in limit), you can click on the volume and click "Live Resize" to increase the limit
mayori
Memory limit for hobby plan is 8GB, 5GB is limit for volume If you're talking about service memory limit, you can go to service settings and increase the replica limit on that page But if you're talking about volume (which has 5GB in limit), you can click on the volume and click "Live Resize" to increase the limit
19 hours ago
He mentioned memory limits in the posted thread.
19 hours ago
I know, but he said 5GB so I think he's talking about volume
2 hours ago
sorry guys, im not that technical, now it shows this error: Description:
I'm experiencing repeated build failures on the Metal builder infrastructure. Three consecutive deployments have failed before Railpack produced any output, with no error logs generated. The service is now offline.
Failed Deployments:
Deployment ID: 8a545d08-51e8-496d-a8b6-bd192b62e3aa (2026-07-15 02:48:33)
Commit: 778582cd (flashcards feature)
Deployment ID: f81892b9-d3bc-4c5f-8919-efcf57e342a0 (2026-07-15 01:19:49)
Commit: 33ae7d3d (reactivar command)
Deployment ID: d7bc0c44-19a4-41ad-9c0a-5b4b182fdeab (2026-07-15 01:16:06)
Commit: 33ae7d3d (reactivar command)
Service Details:
Service ID: 7b999bc3-5469-4717-a1e5-3b5d6f43e000
Service Name: Enarm-bot
Project ID: bf448826-23e5-4274-81d9-6103ea55379b
Runtime: Python 3.11
Repository: floresemiliano2302-lgtm/Enarm-bot (main branch)
Error Pattern: The build fails ~200ms after code snapshot completes, before Railpack executes. No dependency installation logs, no provider detection, no error messages. The source code, dependencies, and configuration are all valid - the same commits previously deployed successfully.
Additional Issue: I also have a memory limit issue: my Hobby plan should provide 5GB, but the service is limited to ~1GB (0.99999744 GB). Please address both issues in this investigation.
Please investigate the Metal builder failure and either repair it or reassign my builds to a functioning builder.
Status changed to Awaiting Railway Response Railway • about 2 hours ago