Hello everyone, I'm reaching out to see if anyone ...
# ask-community
a
Hello everyone, I'm reaching out to see if anyone else is encountering an issue where connecting to port 5432 results in the error: "sqlalchemy.exc.OperationalError: (psycopg2.OperationalError) connection to server at "dagster_postgresql" (172.24.0.2), port 5432 failed: FATAL: sorry, too many clients already" I attempted to resolve this by increasing the maximum number of connections with the following command:
postgres -c 'max_connections=1000'
, but unfortunately, it hasn't resolved the problem. Interestingly, downgrading dagster-postgres seems to mitigate the issue. Has anyone experienced something similar, or does anyone have alternative solutions or insights into this problem? Thank you for any help or advice you can provide.
a
Which DB are you using? AWS RDS / Google CloudSQL / etc ?
a
I'm currently running a local Docker Compose setup with PostgreSQL on my M1 MacBook Pro.
a
mmm strange, how many user-deployments do you have?
a
Hi @Andrea Giardini, Sorry for delay with reply, after many experiments I found out that this is happening because of experimental feature
Global op/asset concurrency
when I turn this off, everything looks okay
a
Can you elaborate a bit? It might be worth investigating it a little more and opening a bug report. Can you post the non-working configuration?