If each deletion event is associated with a unique resource referencing a flow run, it suggests that these events are possibly being triggered by a systematic process or automation that targets individual flow runs. Here are a few things you might consider checking:
1.
Automations and Triggers: Verify if there are any automations or triggers set up in your Prefect environment that might be deleting flow runs based on certain conditions. You can refer to the
Prefect automation triggers documentation for more details on how these work.
2.
Deployment Configurations: Review your deployment configurations, especially any settings related to timeouts, retries, or conditions under which a flow run might be automatically canceled or deleted.
3.
Event Logs: Examine the event logs closely for any patterns or additional metadata that might indicate why each flow run is being deleted. This could include specific error messages, states, or conditions prior to deletion.
4.
Resource Limits: Check if there are any resource limits or quotas in place that might be causing automatic deletions when exceeded. This could be related to the number of concurrent runs, memory usage, etc.
5.
Version-Specific Issues: Ensure that you are using the latest version of Prefect, as some issues might have been resolved in newer releases. You can check the latest release notes for any relevant bug fixes.
If none of these avenues provide clarity, it might be helpful to engage with the Prefect community or support channels for more targeted assistance.