database connection_error ai_generated true

psycopg2.OperationalError: could not connect to server: No such file or directory Is the server running locally and accepting connections on Unix socket "/var/run/postgresql/.s.PGSQL.5432"?

ID: database/postgresql-cannot-connect-to-server-no-such-file

Also available as: JSON · Markdown · 中文
85%Fix Rate
89%Confidence
1Evidence
2024-01-12First Seen

Version Compatibility

VersionStatusIntroducedDeprecatedNotes
PostgreSQL 16.3 active — — —
PostgreSQL 15.7 active — — —
PostgreSQL 14.12 active — — —

Root Cause

PostgreSQL is not running or the Unix socket file is missing, often because the postgres service is stopped, the socket directory is misconfigured, or the server was started with a different socket path.

generic

中文

PostgreSQL 未运行或 Unix 套接字文件丢失，通常是因为 postgres 服务已停止、套接字目录配置错误，或者服务器以不同的套接字路径启动。

Official Documentation

https://www.postgresql.org/docs/16/runtime-config-connection.html

Workarounds

  1. 90% success Check if PostgreSQL is running: systemctl status postgresql (or pg_isready). If not running, start it: systemctl start postgresql (or pg_ctl start -D /var/lib/postgresql/16/main). 
    Check if PostgreSQL is running: systemctl status postgresql (or pg_isready). If not running, start it: systemctl start postgresql (or pg_ctl start -D /var/lib/postgresql/16/main).
  2. 85% success If the socket path is misconfigured, update postgresql.conf to set unix_socket_directories to the correct path (e.g., /var/run/postgresql), then restart: sudo systemctl restart postgresql. 
    If the socket path is misconfigured, update postgresql.conf to set unix_socket_directories to the correct path (e.g., /var/run/postgresql), then restart: sudo systemctl restart postgresql.
  3. 80% success As a workaround, connect via TCP instead of Unix socket by specifying host='localhost' in the connection string: psql -h localhost -U myuser mydb. 
    As a workaround, connect via TCP instead of Unix socket by specifying host='localhost' in the connection string: psql -h localhost -U myuser mydb.

中文步骤

  1. Check if PostgreSQL is running: systemctl status postgresql (or pg_isready). If not running, start it: systemctl start postgresql (or pg_ctl start -D /var/lib/postgresql/16/main).
  2. If the socket path is misconfigured, update postgresql.conf to set unix_socket_directories to the correct path (e.g., /var/run/postgresql), then restart: sudo systemctl restart postgresql.
  3. As a workaround, connect via TCP instead of Unix socket by specifying host='localhost' in the connection string: psql -h localhost -U myuser mydb.

Dead Ends

Common approaches that don't work:

  1. Changing the port in the connection string to a different port (e.g., from 5432 to 5433) 90% fail

    The error is about the socket file, not the port; changing the port will only work if the server listens on that port via TCP, but the socket path remains the same.

  2. Creating a symbolic link to the socket file manually (ln -s /tmp/.s.PGSQL.5432 /var/run/postgresql/.s.PGSQL.5432) 85% fail

    The socket file is created by PostgreSQL at startup; a manual symlink will not be valid if the server is not running, and it may cause permission issues.

  3. Restarting only the application without checking PostgreSQL 100% fail

    The application restart does not affect the database server; the connection will still fail if PostgreSQL is down.