`RuntimeError: Timed out while attempting to conne...
# marvin-ai
j
RuntimeError: Timed out while attempting to connect to ephemeral Prefect API server.
Is this a known issue? We wrap controlflow runs via a retriable mechanism, so it eventually completed. But is this potentially something that is on our end?
Note, this was seen on a brand new environment that we just deployed
j
This is known
Can you tell me if this is on a CI box e.g. something with constrained resources
the server is expected to spin up very quickly (like instantly) so there's not a ton of grace in that timeout period
j
this is running on a Render service 2GB 1CPU
ah i see. we can definitely upgrade it — it's a default setting that we just picked
is there a recommended spec?
j
oh no I think this is set internally
I've been trying to figure out why only CF users seem to see it though
and I'm wondering if its because agents tend to be deployed in lighter envs than say bigger data pipelines
j
I see
mem did spike up to 92%
of our 2GB limit 🙂
maybe it has to be bigger
j
thats the most concrete clue i've heard honestly
a bit surprising (what memory use is actually happening there?) but probably correlated
hm i just checked and the default value for that setting is now 20 seconds
i need to look into this
h
FYI @Michael Napper
We're also seeing this on an 8GB 4 core VM
g
Trying to follow the GCP tutorial but running into the same error `Runtime Error`: https://docs.prefect.io/integrations/prefect-gcp/gcp-worker-guide#step-2-create-a-cloud-run-work-pool Anyone found a work-around?
213 Views