Dan Rammer

[Core feature] Add Container Configuration to Default PodTemplate

1

### Motivation: Why do you think this is important? Currently the default PodTemplate in FlytePropeller allows users to set a base PodSpec configuration for all k8s Pods started by Flyte....

### Motivation: Why do you think this is important? There are a number of configuration values available to the k8s plugin. The Spark plugin constructs a Pods (for Spark executor...

I have the following code testing an INSERT. The goal is to simulate a key already exists; which, when specified with 'ON CONFLICT DO NOTHING' returns no error and instead...

Adding utils for signal ID context

1

# TL;DR With the additional of the signal service in FlyteAdmin we may want to track metrics on a per-signal basis. To do this we need to set the signal...

[Core feature] Add task_execution_deadline configuration to propeller

5

### Motivation: Why do you think this is important? Currenlty, FlytePropeller supports configuration of [node and workflow execution and active deadlines](https://github.com/flyteorg/flytepropeller/blob/287f39269b2b071ca85376af7ebe287d0b5fdd13/pkg/controller/config/config.go#L205-L216). These configuration are used as an upper bound to...

### Motivation: Why do you think this is important? Currently a cached task requires a sequence of state changes to successfully proceed within FlytePropeller. For example, (1) copying inputs from...

### Describe the bug Currently the flyte pod webhook, which is responsible for injecting secrets, [creates a self signed cert](https://github.com/flyteorg/flytepropeller/blob/9cf5368793e5d60113cb819a6e496a7b6accd619/pkg/webhook/init_cert.go#L164) with a 1 year expiration. This is problematic for long-running...

### Describe the bug The [node recovery lookup operations](https://github.com/flyteorg/flytepropeller/blob/8b8e5a8a3db1bdc7a2ad0e66ba78015e6120e97b/pkg/controller/nodes/recovery/client.go#L13-L16) use a node ID to retrieve the most recent node execution. Since task event version 1 these [node IDs are fully...

### Describe the issue FlytePropeller uses the workflowstore to abstract FltyeWorkflow CRD updates and retrievals from the k8s API. To provide additional functionality, this API is designed to wrap other...

[Housekeeping] Enqueue in subqueue only on actionable Pod phase

7

### Describe the issue Currently the FlytePropeller workqueue contains two levels. The first is where FlytePropeller pulls workflows from the periodically evaluate. The second is used as a batching queue...