@Marvin How can we retry a flow run which failed because Prefect was unable to fetch flow code from GitHub?

To retry a failed flow run that didn't successfully fetch code from GitHub, you can make use of Prefect's retry functionality to make the flow more resilient to such transient errors. The most straightforward way to handle retries is to add retry logic directly to the flow or task that fetches the code from GitHub. Here is an example from Prefect documentation that showcases the use of retries for a flow:

python
import httpx
from prefect import flow

@flow(retries=3, retry_delay_seconds=5, log_prints=True)
def get_repo_info(repo_name: str = "PrefectHQ/prefect"):
    url = f"<https://api.github.com/repos/{repo_name}>"
    response = httpx.get(url)
    response.raise_for_status()
    repo = response.json()
    print(f"{repo}")

In the example above, the

retries=3

parameter indicates that the flow should be retried up to 3 times if an error occurs during execution, and

retry_delay_seconds=5

specifies a delay of 5 seconds between each retry. If you need to rerun a flow run that has already failed, you can do so through the Prefect UI or by using the Prefect Client in Python to programmatically create a new flow run. Here's a function from the Prefect documentation that you might use to reschedule failed flow runs:

python
from prefect import get_client
from prefect.client import FlowRunFilter, DeploymentFilter
from prefect.orion.schemas.filters import FlowRunFilter, DeploymentFilter
from prefect.orion.schemas.sorting import FlowRunSort
from datetime import datetime, timezone, timedelta
from prefect.engine.state import Scheduled

async def reschedule_late_flow_runs(deployment_name: str, delay: timedelta, most_recent_n: int, delete_remaining: bool = True, states: Optional[list[str]] = None) -> list[FlowRun]:
    if not states:
        states = ["Late"]
    async with get_client() as client:
        flow_runs = await client.read_flow_runs(
            # Filtering criteria here to identify the failed flow runs
        )
        # The rest of the function logic for rescheduling flows goes here

In this script, you would identify the failed flow runs using appropriate filter criteria and then programmatically reschedule them as needed. For any further details or handling specific cases, you can refer to the Prefect documentation or reach out to the Prefect community via their official channels. Let me know if you need help assembling a more specific script or if you have any other questions concerning retries or handling flow run failures.