On 4 November, access to Railway-hosted applications and railway.com was blocked for some users, not because of anything broken on Railway's own network, but because outside network providers and security vendors were blocking Railway's IP addresses. The disruption ran for 119h 12m, caused entirely by third parties outside Railway's control.
Started
18:54 UTC
Duration
Lasted 4d 23h
Source
IsDown
Next time
Railway is degraded. Is your product working?
An incident on their side does not always mean an outage on yours, and the only way to know is to already be watching your own endpoints. Set that up in under a minute and you will be notified the next time your own site is affected. And there will be a next time.
50 monitors for your own websites. No card.
What happened?
The incident started at 18:54 UTC and was identified the same day by Railway as an external network problem rather than an internal fault. Railway's status page says a third party was sending scanning traffic using forged Railway source IP addresses, and that this spoofed traffic did not originate from Railway's own network. Because of that scanning traffic, some network providers and security services flagged Railway's IP ranges and blocked them, which meant legitimate users on those networks could not reach Railway-hosted applications or the railway.com site itself. Railway explicitly stated this had no connection to the earlier October 21 security incident and that the reports were not evidence of any compromise of Railway services or customer workloads. The first update asked affected users to get their own network administrators to review blocks on Railway's IP addresses, and suggested switching networks as a temporary workaround. A second update at 22:03 UTC confirmed access was still affected on some networks, with review requests submitted to security vendors and no estimated time available for all blocks to be lifted. A third update, timestamped 18:51 UTC, reported that some users had regained access, while Railway continued contacting network operators, security vendors and abuse database providers to resolve what remained. No components are listed as affected in Railway's own breakdown, consistent with this being an external IP reputation and blocking issue rather than a fault in Railway's hosting infrastructure. Only three updates were published across the full 119-hour-plus window, and no user reports were logged in this window on record. No post-mortem has been published, and the record does not show a final resolution update closing the incident out.
Learning
This incident sits on your network path, not inside your application or inside Railway's hosting layer, which is exactly why it is easy to misdiagnose from your side. A third party forging your provider's IP addresses to send scanning traffic can get those IP ranges blocked or downgraded in reputation by security vendors and ISPs who have no direct relationship with Railway at all, and that blocking happens on networks Railway does not operate and cannot unilaterally fix. Your own uptime checks against your application endpoint may show nothing wrong, because the application itself is healthy, while users on specific blocked networks still cannot reach you at all. That is the gap this kind of outage exposes: a check that only tests reachability from one vantage point will miss blocks that only apply on certain other networks. Railway's median incident over the last 90 days resolved in 1 hour 32 minutes, against 36 incidents, so an incident running past 119 hours with only three published updates is unusual for them, and Railway's own channel is the only place tracking whether individual network blocks have actually been lifted. Watching from multiple external vantage points, rather than just your own backend, is the only way to see a problem like this as it is actually distributed. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.