Hey Prefect Folks!
Using Prefect 1.2, I wanted to use the server on a VM (Fedora 35) to deploy my flows. It all runs through relatively smoothly, but then I get the following errors:
Pulling postgres ... done
Pulling hasura ... done
Pulling graphql ... done
Pulling apollo ... done
Pulling towel ... done
Pulling ui ... done
Creating tmp_postgres_1 ...
Creating tmp_postgres_1 ... error
ERROR: for tmp_postgres_1 Cannot start service postgres: driver failed programming external connectivity on endpoint tmp_postgres_1 (***): Error starting userland proxy: listen tcp4 127.0.0.1:5432: bind: address already in use
ERROR: for postgres Cannot start service postgres: driver failed programming external connectivity on endpoint tmp_postgres_1 (***): Error starting userland proxy: listen tcp4 127.0.0.1:5432: bind: address already in use
I have a local Postgres instance that has to be there and must not be killed. What would be my options here to have the server running? I considered port forwarding, but I am not sure how to do that.
Any help is appreciated
looks like an issue with your DB drivers in the container - perhaps try a more recent Postgres version?
are you just getting started with Prefect? If so, it’s better to start with Prefect 2.0, which is also much easier to set up and use:
Short answer: we highly encourage all new Prefect users to start with
Prefect 2.0 and Cloud 2.0.
At the time of writing, both Prefect 2.0 and Cloud 2.0 will remain under active development in beta status until approximately July 2022 , with a number of major features yet to come.
The target for the General Availability of Cloud 2.0 and stable release of Prefect 2.0 is July 2022.
Can I use Cloud 2.0 in production already?
Cloud 2.0 is a beta product, with beta product guarantees. When decid…
You are most likely right. But naturally I prefer clean solutions than tinkering with lower-level orchestration stuff. So I will give Prefect Orion another shot
I switched back to Prefect 1.2, because of my issues here:
I recently switched to Prefect Orion for a project of mine.
Since I had used Prefect 1.0 (with pleasure) before, I thought I could refactor my data pipeline. I planned to connect to the database and pass around the connection object as was possible with Prefect 1.0.
So I defined my pipeline as follows (please note that I omitted some more sensitive information):
import sqlalchemy as sql
from prefect import task, flow
from prefect.task_runners import SequentialTaskRu…
However, Prefect 2 is indeed easier to set up and use (not least because of SQLite as the default), so I will rewrite my flows to accomodate the issues I mentioned in the topic above.
Sounds great! FWIW, I did create a GitHub issue for the DB connection issue you mentioned, so this should be tackled in a near future