Railway Outage History

Started May 23, 2026, 04:11 PM UTC· resolved May 23, 2026, 07:45 PM UTC

1. identified May 23, 2026, 04:11 PM UTC

   GitHub is experiencing intermittent errors with app installation token authentication: https://www.githubstatus.com/incidents/k5z4d1v1tqmt. This may cause "Bad credentials" or "repository not authorized" errors when deploying from GitHub. Deploys should recover as GitHub resolves this upstream issue.

2. identified May 23, 2026, 05:06 PM UTC

   GitHub continues to investigate elevated authentication failures for app installation tokens. Deploys from GitHub may intermittently fail with "Bad credentials" errors. We're monitoring and will update when GitHub resolves this.

3. identified May 23, 2026, 05:47 PM UTC

   GitHub is seeing signs of mitigation and is continuing to monitor for complete mitigation.

4. monitoring May 23, 2026, 07:05 PM UTC

   GitHub is seeing significant signs of mitigation and monitoring for full mitigation.

5. resolved May 23, 2026, 07:45 PM UTC

   The GitHub incident has been resolved.

Read the full incident report →

Started May 22, 2026, 02:45 PM UTC· resolved May 22, 2026, 03:07 PM UTC

1. investigating May 22, 2026, 02:45 PM UTC

   We are investigating reports of intermittent connectivity issues affecting railway.com and user-deployed services. Some users may have experienced SSL handshake failures, 400 errors, or brief unavailability when accessing the platform or their workloads. We are actively investigating the root cause and will provide updates as we learn more.

2. monitoring May 22, 2026, 03:02 PM UTC

   We have identified the cause of the connectivity issues and have deployed a fix that is being The affected workload has been isolated and connectivity has recovered. We are continuing to monitor to confirm stability

3. resolved May 22, 2026, 03:07 PM UTC

   We have identified the cause of the connectivity issues and deployed a fix. Connectivity has fully recovered. We apologize for any disruption this caused to your services. Update: Following the initial resolution, a brief spike in HTTP request latency was observed in the EU West region for approximately 1-2 minutes. Users in this region may have experienced slower response times. The issue has subsided.

Read the full incident report →

Started May 20, 2026, 02:25 AM UTC· resolved May 20, 2026, 07:57 AM UTC

1. investigating May 19, 2026, 10:29 PM UTC

   We are investigating a widespread service disruption affecting Railway. Users may be experiencing errors including "no healthy upstream", "unconditional drop overload", login failures, and inability to access the dashboard. We are actively investigating and will provide updates as soon as possible.

2. identified May 19, 2026, 10:43 PM UTC

   We have identified the cause of the disruption. Access to our upstream cloud provider has been restored and we are working on a fix. The Railway dashboard and services running on our cloud infrastructure may continue to be affected until the fix is deployed.

3. identified May 19, 2026, 11:37 PM UTC

   Google Cloud has blocked our account, making some Railway services unavailable. We have escalated this directly with Google. The Railway Platform team has since confirmed access to Google Cloud and is working on restoring access to all workloads. We have access to some of our Google Cloud–hosted infrastructure and are working to restore the rest of the service. We apologize for the disruption.

4. identified May 20, 2026, 12:37 AM UTC

   We are working to restore the Google Cloud infrastructure that powers our dashboard, API, and internal network's control plane. We are in direct contact with Google Cloud's support team. We do not have an ETA at this time.

5. identified May 20, 2026, 01:23 AM UTC

   We are continuing to work on restoring access to our workloads on Google Cloud. In parallel, our infrastructure team is evaluating alternative paths to bring affected services back online. We do not have an ETA at this time and will continue posting updates as we have them.

6. identified May 20, 2026, 01:34 AM UTC

   We have recovered our compute on Google Cloud, but services are unable to start because of ongoing networking issues on Google Cloud's side. We are engaged with Google Cloud support to resolve this and will post the next update as soon as we have more to share.

7. identified May 20, 2026, 01:41 AM UTC

   We are seeing gradual recovery on Railway metal workloads. To ensure things remain stable as we ramp back up, we are temporarily throttling all non-enterprise builds to avoid overwhelming our build infrastructure.

8. identified May 20, 2026, 03:05 AM UTC

   More workloads are coming back online. Some users may still experience intermittent issues during the recovery. Non-enterprise deploys remain paused; enterprise deploys are unaffected.

9. identified May 20, 2026, 04:58 AM UTC

   Deploys are now possible. Workloads still hosted on Google Cloud may continue to experience intermittent issues while we complete recovery. For ongoing context, see this FAQ article: https://station.railway.com/community/what-we-know-so-far-may-19th-2026-86354cdd. If you need direct support, please open a thread at https://station.railway.com.

10. monitoring May 20, 2026, 06:14 AM UTC

    Railway services have fully recovered. Some workloads may still need a redeploy, we're automatically redeploying any we detect as unhealthy. If your service isn't responding correctly, please trigger a redeploy from the dashboard or CLI. We're sorry for the disruption. A detailed postmortem will follow once we've confirmed stability.

11. resolved May 20, 2026, 07:57 AM UTC

    We have resolved this incident and a post mortem is available here. https://blog.railway.com/p/incident-report-may-19-2026-gcp-account-outage

Read the full incident report →

Started May 19, 2026, 08:44 PM UTC· resolved May 19, 2026, 08:45 PM UTC

1. investigating May 19, 2026, 08:20 PM UTC

   Between approximately 20:20 and 20:25 UTC, we experienced a brief disruption to our compute infrastructure in US East. Existing running services were not affected but during this window, users may have been unable to deploy or perform actions in this region. The issue was resolved within approximately 5 minutes and all operations returned to normal.

2. resolved May 19, 2026, 08:25 PM UTC

   The issue was resolved within approximately 5 minutes and all operations returned to normal.

Read the full incident report →

Started May 15, 2026, 02:11 PM UTC· resolved May 15, 2026, 02:52 PM UTC

1. investigating May 15, 2026, 02:11 PM UTC

   We are currently investigating this incident

2. identified May 15, 2026, 02:22 PM UTC

   We have identified the issue and are working towards a fix.

3. monitoring May 15, 2026, 02:38 PM UTC

   We are seeing recovery and are monitoring the results. Metric and log request times should be at normal latency

4. resolved May 15, 2026, 02:52 PM UTC

   We have seen sustained recovery and normal latency. This incident has been resolved.

Read the full incident report →

Started May 08, 2026, 08:09 PM UTC· resolved May 08, 2026, 09:14 PM UTC

1. monitoring May 08, 2026, 08:09 PM UTC

   We are aware of an issue affecting SSL certificate issuance for custom domains. Users adding or renewing custom domains may see certificates fail to provision. This is caused by an upstream incident with our certificate provider. We are monitoring the situation and will provide updates.

2. resolved May 08, 2026, 09:14 PM UTC

   The upstream issue affecting SSL certificate issuance has been resolved. If your domain shows a "Failed to issue TLS certificate" error, click "Try Again" to re-issue the certificate.

Read the full incident report →

Started May 06, 2026, 10:52 AM UTC· resolved May 06, 2026, 12:03 PM UTC

1. investigating May 06, 2026, 10:52 AM UTC

   We are aware of intermittent connection issues affecting traffic in our EU West region. Users may experience connection resets (ECONNRESET) or failed requests when reaching services in this region. We are actively investigating and we will provide updates as we learn more.

2. monitoring May 06, 2026, 11:16 AM UTC

   We have identified a single unhealthy proxy in our EU West region as the source of the connection issues. The affected proxy has been removed from rotation and traffic is being served by the remaining healthy proxies. We are monitoring the region for full recovery. Users should see connections returning to normal.

3. resolved May 06, 2026, 12:03 PM UTC

   This incident is resolved

Read the full incident report →

Started May 04, 2026, 07:41 PM UTC· resolved May 04, 2026, 08:06 PM UTC

1. investigating May 04, 2026, 07:41 PM UTC

   We are aware of an issue affecting edge routing in our Southeast Asia (Singapore) region. Users may experience failed requests or DNS resolution errors when reaching services deployed in this region. We are actively investigating and will provide updates as we learn more.

2. monitoring May 04, 2026, 07:50 PM UTC

   Edge routing has recovered in Southeast Asia and we are monitoring the results.

3. resolved May 04, 2026, 08:06 PM UTC

   We have seen consistent recovery and are marking this incident as resolved.

Read the full incident report →

Started May 04, 2026, 05:21 PM UTC· resolved May 04, 2026, 05:46 PM UTC

1. investigating May 04, 2026, 05:21 PM UTC

   We are aware of elevated latency affecting our edge network in US East. Requests to services running in this region may take longer than normal to complete. We are actively investigating and we will provide updates as we learn more.

2. identified May 04, 2026, 05:34 PM UTC

   We have identified the issue affecting US East edge latency and are rolling out a fix. Some users may already see improvement, though latency remains elevated for others. We will continue to provide updates as the fix takes effect.

3. monitoring May 04, 2026, 05:39 PM UTC

   We suspect the elevated latency is related to an issue with our upstream CDN layer. As a mitigation, we have removed the affected provider from our edge network and are now seeing performance return to normal. We will continue to monitor before resolving.

4. resolved May 04, 2026, 05:46 PM UTC

   The elevated latency affecting our US East edge network has been resolved. We removed the affected upstream provider, and performance has returned to normal. We apologize for any inconvenience. If you continue to experience issues, please contact us at https://station.railway.com.

Read the full incident report →

Started May 04, 2026, 03:38 PM UTC· resolved May 04, 2026, 04:50 PM UTC

1. investigating May 04, 2026, 03:38 PM UTC

   We are investigating degraded performance affecting stateful services (services with attached volumes). Affected services may experience elevated latency or slower disk I/O than expected. Stateless services and new deployments without volumes are not affected.

2. monitoring May 04, 2026, 04:14 PM UTC

   We have implemented a fix and are monitoring the results. Performance is back to a healthy state.

3. resolved May 04, 2026, 04:50 PM UTC

   We have seen consistently healthy stateful service performance. This incident has been resolved.

Read the full incident report →

Started May 01, 2026, 06:26 PM UTC· resolved May 01, 2026, 06:27 PM UTC

1. monitoring May 01, 2026, 06:26 PM UTC

   Some users were unable to log in to Central Station for a brief period of approximately five minutes during 1800 ~ 1805 UTC. A fix has been deployed and we are monitoring the situation.

2. resolved May 01, 2026, 06:27 PM UTC

   This incident has been resolved.

Read the full incident report →

Started Apr 29, 2026, 10:48 AM UTC· resolved Apr 29, 2026, 11:33 AM UTC

1. investigating Apr 29, 2026, 10:48 AM UTC

   Our CDN provider Fastly is experiencing an outage, which is causing TLS provisioning to take longer than expected. As a result, new domains being added to Railway might fail to obtain SSL certificates. Fastly's incident is available here: https://www.fastlystatus.com/incident/378511

2. monitoring Apr 29, 2026, 11:10 AM UTC

   Fastly has implemented a fix and is experiencing recovery. They are currently monitoring this incident. We see active recovery in TLS provisioning globally. Please retry adding your domain. If you have any issues contact us on https://station.railway.com

3. resolved Apr 29, 2026, 11:33 AM UTC

   We have seen full recovery and are resolving this incident. If you have any questions please reach out at https://station.railway.com.

Read the full incident report →

Started Apr 24, 2026, 03:34 PM UTC· resolved Apr 24, 2026, 04:03 PM UTC

1. investigating Apr 24, 2026, 03:34 PM UTC

   We are aware that the Dashboard (railway.com) may be currently unavailable. Users may not be able to access the dashboard to manage their projects and services. All existing deployments and running services are not affected. We are actively investigating and will provide updates as we learn more.

2. monitoring Apr 24, 2026, 03:39 PM UTC

   We are seeing recovery of the dashboard.

3. resolved Apr 24, 2026, 04:03 PM UTC

   The dashboard is fully available.

Read the full incident report →

Started Apr 23, 2026, 02:56 PM UTC· resolved Apr 23, 2026, 03:32 PM UTC

1. identified Apr 23, 2026, 02:56 PM UTC

   GitHub - https://www.githubstatus.com/incidents/3f30ycplyr7l

2. resolved Apr 23, 2026, 03:32 PM UTC

   GitHub has resolved this incident.

Read the full incident report →

Started Apr 13, 2026, 03:13 PM UTC· resolved Apr 13, 2026, 03:32 PM UTC

1. investigating Apr 13, 2026, 03:13 PM UTC

   We have temporarily paused deployments for Trial tier users to maintain platform stability. Hobby and Pro users are unaffected. We will provide an update once deployments are restored for all tiers.

2. monitoring Apr 13, 2026, 03:22 PM UTC

   Deployments are now enabled for all tiers and the queue has returned to normal. We are continuing to monitor to ensure stability.

3. resolved Apr 13, 2026, 03:32 PM UTC

   Automatically resolved.

Read the full incident report →

Started Apr 10, 2026, 11:56 AM UTC· resolved Apr 10, 2026, 12:01 PM UTC

1. investigating Apr 10, 2026, 11:56 AM UTC

   We are experiencing degraded access to the Railway dashboard.

2. resolved Apr 10, 2026, 12:01 PM UTC

   We experienced a brief period of degraded access to the Railway dashboard due to elevated traffic. We enabled additional protection and access was restored. Running services, deployments, and the API were not affected. We apologize for any inconvenience.

Read the full incident report →

Started Apr 08, 2026, 11:53 PM UTC· resolved Apr 09, 2026, 12:12 AM UTC

1. investigating Apr 08, 2026, 11:53 PM UTC

   We're aware of an issue causing the Railway dashboard to load slowly or fail to load for some users. Our team is actively investigating the root cause.

2. identified Apr 08, 2026, 11:59 PM UTC

   We've identified the source of the degraded dashboard performance and are working on a fix.

3. monitoring Apr 09, 2026, 12:03 AM UTC

   A fix has been deployed. All systems appear to be operating normally. We are monitoring dashboard performance.

4. investigating Apr 09, 2026, 12:08 AM UTC

   The dashboard is back to normal performance. The issue has been fully resolved and we've confirmed stability through monitoring.

5. resolved Apr 09, 2026, 12:12 AM UTC

   This incident has been resolved

Read the full incident report →

Started Apr 06, 2026, 03:09 PM UTC· resolved Apr 06, 2026, 04:26 PM UTC

1. investigating Apr 06, 2026, 03:09 PM UTC

   We are currently investigating this issue.

2. investigating Apr 06, 2026, 03:17 PM UTC

   Triage escalated to incident

3. identified Apr 06, 2026, 03:17 PM UTC

   We are currently investigating this issue

4. identified Apr 06, 2026, 03:40 PM UTC

   We have paused hobby deploys to alleviate deploy pressure.

5. monitoring Apr 06, 2026, 04:21 PM UTC

   We have unpaused hobby as we are seeing recovery. We are currently monitoring this

6. resolved Apr 06, 2026, 04:26 PM UTC

   This incident is resolved

Read the full incident report →

Started Apr 02, 2026, 01:50 PM UTC· resolved Apr 02, 2026, 03:03 PM UTC

1. investigating Apr 02, 2026, 01:50 PM UTC

   We are currently investigating reports of canvas load failures and slow builds.

2. investigating Apr 02, 2026, 02:15 PM UTC

   We're seeing recovery across affected systems. Canvas loading and build times are returning to normal levels. Our team is continuing to investigate the root cause to ensure full resolution. We'll post another update once we've confirmed stability. If you're still experiencing issues, please try refreshing your dashboard. We appreciate your patience.

3. monitoring Apr 02, 2026, 02:31 PM UTC

   We've identified the root cause as abnormally high database load on an internal service, which was impacting canvas loading and build performance. The load has subsided and systems are recovering on their own. We'll confirm once we've verified sustained stability. Thank you for your patience.

4. resolved Apr 02, 2026, 03:03 PM UTC

   This incident has been resolved. Canvas loading and build performance are back to normal and have been stable. Thank you for your patience while we worked through this.

Read the full incident report →

Started Mar 30, 2026, 12:20 PM UTC· resolved Mar 30, 2026, 05:37 PM UTC

1. investigating Mar 30, 2026, 12:20 PM UTC

   While updating our edge CDN to allow us to identify cached assets on a per-domain basis, we enabled a feature called “Surrogate Keys”. Unknowingly to us, this bypassed our “if CDN is disabled for this domain, pass directly to origin” behavior, which meant that we accidentally enabled CDN caching capabilities for domains that didn’t previously have it enabled. Because of that, the following assets may have been cached during this 30 minute period even with domains with the CDN feature disabled: Any content with public Cache-Control headers Images, videos and other known-static assets No HTML would’ve been cached; this is always considered dynamic At 11:30 UTC, once realizing the mistake, we immediately pushed an update which reverted this behavior and purged the cache of all assets globally.

2. monitoring Mar 30, 2026, 12:21 PM UTC

   This is a retroactive incident that was resolved at 11:30 UTC. We are still determining impact.

3. monitoring Mar 30, 2026, 12:27 PM UTC

   We are seeing reports that even content without cache-control set was erroneously cached. We are still determining impact.

4. resolved Mar 30, 2026, 05:37 PM UTC

   This incident has been resolved. A postmortem is available at https://blog.railway.com/p/incident-report-march-30-2026-accidental-cdn-caching.

Read the full incident report →

Started Mar 26, 2026, 08:46 PM UTC· resolved Mar 26, 2026, 11:27 PM UTC

1. investigating Mar 26, 2026, 07:50 PM UTC

   Some users in or around Europe may be experiencing increased latency when accessing their workloads.

2. monitoring Mar 26, 2026, 10:54 PM UTC

   We have made some configuration changes and are seeing improvements.

3. resolved Mar 26, 2026, 11:27 PM UTC

   We have witnessed full recovery, and response times are back to their nominal values.

Read the full incident report →

Started Mar 26, 2026, 07:38 AM UTC· resolved Mar 26, 2026, 07:41 AM UTC

1. investigating Mar 26, 2026, 07:38 AM UTC

   The dashboard is failing to load with the message "The train was derailed".

2. identified Mar 26, 2026, 07:39 AM UTC

   We have identified the issue and are rolling out a fix.

3. resolved Mar 26, 2026, 07:41 AM UTC

   The dashboard is fully accessible again.

Read the full incident report →

Started Mar 25, 2026, 08:23 PM UTC· resolved Mar 25, 2026, 09:47 PM UTC

1. investigating Mar 25, 2026, 08:23 PM UTC

   We are currently investigating this incident.

2. identified Mar 25, 2026, 09:17 PM UTC

   We have traced this back to a change rolled out with static IP egress gateways. We have begun the rollbacks and are already seeing recovery.

3. monitoring Mar 25, 2026, 09:29 PM UTC

   The rollback has completed, we are seeing recovery, and are currently monitoring the result.

4. resolved Mar 25, 2026, 09:47 PM UTC

   This incident has been resolved.

Read the full incident report →

Started Mar 25, 2026, 06:05 PM UTC· resolved Mar 25, 2026, 07:21 PM UTC

1. investigating Mar 25, 2026, 06:05 PM UTC

   Dockerhub Registry Outage, pulls from Dockerhub may fail..

2. resolved Mar 25, 2026, 07:21 PM UTC

   Dockerhub reports recovery.

Read the full incident report →

Started Mar 25, 2026, 04:09 PM UTC· resolved Mar 25, 2026, 05:54 PM UTC

1. investigating Mar 25, 2026, 04:09 PM UTC

   Cert discrepancy between Fastly and Railway leading to SAN errors. The network team is actively triaging the issue for affected customers..

2. identified Mar 25, 2026, 04:11 PM UTC

   For some affected domains, cert provisioning is inoperable, this is due to incorrect domain configuration, please re-add your domains to triage the issue..

3. identified Mar 25, 2026, 04:38 PM UTC

   Manual workarounds for customers show that access is restored, we are working with our upstream vendor to implement domain forwarding to mitigate customer misconfiguration..

4. resolved Mar 25, 2026, 05:54 PM UTC

   Network behavior is back to nominal..

Read the full incident report →