Railway outages have made it hard to access projects and deploy updates, and reliability has been a recurring concern. What would you consider moving to, and how would you make a move less disruptive?
What are some more reliable alternatives to Railway?
Answers
I would look at a different hosting provider, but check its underlying cloud dependencies too. Moving to a service that relies on the same infrastructure may not protect you from the same kind of outage.
Compare incident histories, support options, deployment features and pricing against what your projects actually need. The discussion here doesn't establish a particular provider as more reliable.
Reply to this answer
Make the deployment portable before picking the next host. Keep your container images in a registry you control and maintain a Compose or Kubernetes setup that can run on another provider.
A short DNS TTL can help when switching traffic, but it isn't a complete failover plan. You still need a working second deployment and a plan for your data and configuration.
Reply to this answer
First figure out whether the immediate problem is that the app is down or that you can't deploy changes. Those call for different preparations. A running app may stay available while the deployment system is inaccessible, but that won't help if an urgent fix needs to go out.
Reply to this answer
If these are production services, treat inaccessible support as part of the risk, not just the deployment outage. Check what support channels and response commitments a prospective host offers, and decide what you'll do when you can't reach them.
Reply to this answer
I wouldn't assume every outage has the same cause. The discussion raises possible connections to wider certificate and PHP problems, but those are guesses, not a confirmed explanation for Railway's issues.
Reply to this answer
Add your answer
Answers are reviewed before publication.
9g+claspquartz_261:
Testing that second deployment matters too. A setup that only exists on paper can be rough to debug during an outage.