New RunLLM response generated: >>>*# Question* Could anyone comment on the queue implementation in Flyte? Are those queues durable? What happens to the queue state in case of flyte-binary unexpected restart? Are there any settings we can change with respect to queue durability? # Answer Flyte uses Kubernetes and etcd to maintain state, ensuring durability and recoverability. FlytePropeller, the core engine, maintains state in etcd, which allows it to survive restarts and continue processing workflows (source). Flyte's design ensures that even if the control plane restarts, the system can recover and continue execution without losing progress (source). For queue durability, FlytePropeller's state is stored in etcd, making it resilient to failures. However, specific settings for queue durability are not explicitly mentioned in the provided data sources. Flyte's retry mechanisms and state management in etcd ensure that tasks and workflows can recover from failures (source). Would you like more detail on any specific aspect? Sources: - https://discuss.flyte.org/t/13252961/hello-does-anyone-know-what-s-the-expected-behaviour-when-fl/ - https://discuss.flyte.org/t/10116097/hi-i-m-evaluating-flyte-for-our-workflow-orchestration-needs/ - https://flyte.org/blog/build-indestructible-pipelines-with-flyte Slack Link https://flyte-org.slack.com/archives/CP2HDHKE1/p1725524210.463329 👍 0 button 👎 0 button 💡 Hint: Mention @runllm in the thread for followups.