Dagster had 5 outages in the last 2 years totaling 14h 3m of downtime — averaging 0.2 incidents per month.
There were 5 Dagster outages since June 27, 2025 totaling 14h 3m of downtime. Each is summarised below — incident details, duration, and resolution information.
We observed an increase in Dagster+ serverless runs failing to start between 10:50AM and 11:00AM PST. Our cloud provider indicated an incident on their end which we have linked to these issues. Failure rate has since gone back down, we will post more information as we learn more.
The underlying issue has been resolved.
A configuration change caused an delay in scheduling runs and sensors. We identified and fixed the issue. The issue was identified at 12:15pm PST and was resolved at 1:30pm PST.
We're investigating an issue where some users are seeing a version of the old Dagster+ UI. We've identified the root cause of the problem and are investigating a fix.
This issue has now been resolved - all users should now be seeing the new Dagster+ UI again.
We observed some DNS resolution related API latency and errors from 10:10pm-10:35pm EST. We've identified the issue and have released a fix.
We observed some API latency and errors from 2:15pm-2:45pm EST. We've identified the issue and have released a fix.
This issue has now been resolved.