Prefect
Apps and SaaS. Source: Prefect status page.
- Incidents, last 90 days
- 6
- Median duration
- 14m
- p95 duration
- 1h 44m
- Naming an upstream provider
- 16%
- With a postmortem on the status page
- 0%
What failed, last 90 days
| A provider or third-party service this vendor depends on failed | 3 |
| The text does not say enough to tell | 1 |
| Network, DNS, routing or connectivity failure | 1 |
| Exhausted memory, disk, CPU, queues, connection pools or quotas | 1 |
Tagged by claude-haiku-5-5 from each incident's updates. See the methodology.
Incidents per month
Hatched months are before the recorded history starts.
All recorded incidents
Flow Runs Crashing on Modal Push Work Pools
Failure mode: A provider or third-party service this vendor depends on failed. Root cause: A failure at a provider or vendor they depend on. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).
-
identified
Since 18:00 UTC on October 16, most flow runs on Modal push work pools crash during infrastructure startup. Affected runs move from Pending to Crashed with the message "sandbox exited with non-zero code 137." We have tracked the issue to an upgrade in the Modal client library...
-
resolved
This incident has been resolved and Modal push pools are fully operational.
Excerpts are shortened. Read the full incident on Prefect's status page.
Elevated API Error Rates
Failure mode: The text does not say enough to tell. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).
-
resolved
Between 11:08 and 11:16 AM EDT, a subset of requests to the Prefect Cloud API returned elevated error rates. Some users may have seen failed or slow API requests and delays in background processing during this window. The issue has been resolved and all services are operating...
Excerpts are shortened. Read the full incident on Prefect's status page.
Elevated connection errors
Failure mode: Network, DNS, routing or connectivity failure. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).
-
monitoring
Between 19:17 and 19:25 UTC, some real-time connections to the Prefect Cloud UI and event streams were interrupted, causing a brief period of elevated errors. Affected clients reconnected automatically.
-
resolved
This incident has been resolved.
Excerpts are shortened. Read the full incident on Prefect's status page.
Prefect Serverless service disruption
Failure mode: A provider or third-party service this vendor depends on failed. Root cause: A failure at a provider or vendor they depend on. Tagged by claude-haiku-5-5 with high confidence (taxonomy v0.2).
Names an upstream provider: AWS.
-
resolved
Between about 22:49 and 00:06 UTC on 21 September, AWS could not start new EC2 and Fargate containers in the `us-east-1` region. This caused flow runs submitted in that window to crash. The issue has since been resolved and Prefect Serverless is operating normally.
Excerpts are shortened. Read the full incident on Prefect's status page.
Service disruption for outbound email
Failure mode: A provider or third-party service this vendor depends on failed. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with low confidence (taxonomy v0.2).
-
identified
We've identified the root cause of the issue affecting outbound email. Customers signing in with an email magic link may be unable to log in. Email invitations and automation email notifications are delayed. Flow runs and API access are unaffected. We'll post another update when...
-
monitoring
A fix has been implemented and we're monitoring the results. Email services should now be operating normally. We'll continue to watch closely to confirm full recovery before marking this incident as resolved.
-
resolved
This incident has been resolved. Email-related services are fully operational. We apologize for any inconvenience and thank you for your patience.
Excerpts are shortened. Read the full incident on Prefect's status page.
Service disruption for logs component
Failure mode: Exhausted memory, disk, CPU, queues, connection pools or quotas. Root cause: A bug in the vendor's own code. Tagged by claude-haiku-5-5 with medium confidence (taxonomy v0.2).
-
investigating
We're currently investigating an issue affecting our logs service. Users may experience issues viewing logs. Our team is actively looking into the cause, and we'll provide an update as soon as we have more information.
-
identified
We've identified the root cause and are working on a fix. Logs may continue to be affected until the fix is fully rolled out. We'll post another update when the issue is resolved.
-
monitoring
A fix has been implemented and we're monitoring the results. Logs should now be operating normally. We'll continue to watch closely to confirm full recovery before marking this incident as resolved.
-
resolved
This incident has been resolved. Logs are fully operational. The issue was caused by an edge case in autoscaling, and we've taken steps to prevent recurrence. We apologize for any inconvenience and thank you for your patience.
Excerpts are shortened. Read the full incident on Prefect's status page.
Elevated errors and degraded orchestration performance
Failure mode: Orchestration, control plane, service discovery or load balancing failed. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with low confidence (taxonomy v0.2).
-
investigating
_We're investigating elevated errors and degraded orchestration performance affecting Prefect Cloud (US). Some API requests may return errors and flow run scheduling may be delayed. We're actively working to resolve this and will post updates as we learn more._
-
monitoring
_The underlying database issue has been resolved and the system is recovering. API error rates are dropping and flow run scheduling is catching up. Some automations and event processing may remain delayed while a backlog clears._
-
resolved
The issue affecting the Prefect Cloud API and UI has been fully resolved and all services are operating normally, including background processing. We apologize for the disruption.
Excerpts are shortened. Read the full incident on Prefect's status page.
Github incident is impacting Prefect Cloud managed execution and Prefect Horizon operations
Failure mode: A provider or third-party service this vendor depends on failed. Root cause: A failure at a provider or vendor they depend on. Tagged by claude-haiku-5-5 with high confidence (taxonomy v0.2).
-
monitoring
This is an advisory incident, please track the status of Github at [https://www.githubstatus.com](https://www.githubstatus.com/)/.
-
resolved
As of 9:25pm ET, we're seeing error rates and latency recover on Github-related operations. Although their incident is still active, the impact on Prefect Cloud is resolved.
Excerpts are shortened. Read the full incident on Prefect's status page.
Elevated Error Rates
Failure mode: The text does not say enough to tell. Root cause: The vendor does not say why it happened. Tagged by claude-haiku-5-5 with low confidence (taxonomy v0.2).
-
identified
We are investigating an issue causing intermittent errors when loading flow run lists and details in the UI and API. Other Prefect Cloud functionality - including flow run execution, scheduling, and task orchestration - is not affected, and your workflows continue to run...
-
resolved
Between roughly 15:50 and 16:15 UTC, some requests intermittently returned errors. Execution of flows and tasks was unaffected. Service returned to normal by ~16:15 UTC.
Excerpts are shortened. Read the full incident on Prefect's status page.