Hey folks - I'm struggling to handle prefect flow pod evictions elegantly - wondering if anyone here has found good solution. I've updated to the most recent prefect-kubernetes (0.3.2) for my worker which now crashes evicted flows instead of letting them run indefinitely, but I am having problems with a retry.
I think these two issues make it difficult to do this well:
• Support retrying flows on crashed state: https://github.com/PrefectHQ/prefect/issues/10211
• Add an Automations action to Retry a flow run: https://github.com/PrefectHQ/prefect/issues/11431
Is there a better way that I am missing?
Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.