most-monitor-27728
11/14/2024, 10:37 PMbulky-afternoon-92433
11/15/2024, 8:57 AM/flows/{flow_id}
therefore gets all runs for a flow in one go, which depending on the deployment can be a massive set. I suspect this is why you're experiencing timeouts as well.
There is an easy workaround for your case though as you're accessing a specific run. You should be able to forego the runs listing completely by instantiating a run directly with a pathspec:
from metaflow import Run
run = Run("FlowName/run_id")
bulky-afternoon-92433
11/15/2024, 9:01 AMbulky-afternoon-92433
11/15/2024, 9:06 AMDBConfiguration
class. You can set the environment variable
MF_METADATA_DB_TIMEOUT=<your-value>
for the deployment and that should be picked up.most-monitor-27728
11/15/2024, 9:18 PMMF_METADATA_DB_TIMEOUT
envvar. I set it up and it works as expected. Maybe it could be added to the documentation to make it more visible.
Thanks again.