Railway - Notice history

(This is a retroactive incident) During this time we noticed a single host in the US East 1 region had restarted and we quickly worked to bring containers back online, users may have not been able to dial their deployments or connect to databases during the restart efforts. This incident has been resolved.

This is a retroactive incident which is resolved. For the period of the incident (11:45-12:35 UTC), some workloads may have exhibited connection problems. The incident was due to resource exhaustion on the two hosts due to unaccounted resource thresholds. We have improved our monitoring to better respond to this type of incident and are working on preventative measures in our orchestration.

We have observed degraded performance of the edge proxy layer. Users may experience dropped connections. We are investigating.

We implemented a fix and are currently monitoring the result.

A single host in the US West 1 region has restarted and services are coming back online.

This is a retroactive incident about severe service degradation that happened between 12:40 PM UTC and 13:10 PM UTC on our edge proxies in south-east Asia. We are conducting an investigation into what happened and will update this incident once it's complete.

Our new edge proxy is experiencing a major outage. We have identified the issue and are working on a fix

Our proxy in Europe continued to process requests but from August 25th, 20:30 to August 26th, 15:00, HTTP latency may have been elevated as we dealt with a sudden increase in traffic from this region.