Instead of using /usr/local/pgsql/data, I create my database cluster with the following code
initdb -D /tmp/psql
pg_ctl -D /tmp/psql -l logfile -o "--unix_socket_directories='$PWD'" start
But when I run psql, I get the following error.
psql: error: could not connect to server: No such file or directory
Is the server running locally and accepting
connections on Unix domain socket "/tmp/.s.PGSQL.5432"?
I have read the doc and I cannot find related options for changing the location for database cluster. What have I missed?
I ran into the same issue and specifying the socket directory with the -h option fixes the issue.
psql -h $PWD -p 5432 postgres
Related
I've installed posgresql:
nix-env -iA nixos.postgresql
Now I wanto use it.
psql
psql: error: connection to server on socket "/run/postgresql/.s.PGSQL.5432" failed: No such file or directory
Is the server running locally and accepting connections on that socket?
Must I add something in /etc/nixos/configuration.nix ?
update
I've tried to do like this link. i.e. install in nix-shell and not in nix-env.
I've the same shell.nix
nix-shell --pure shell.nix
done
server started
psql
psql: error: could not connect to server: could not connect to server: No such file or directory
Is the server running locally and accepting
connections on Unix domain socket "/mnt/c/Users/Pierre-Olivier/nix/psql/.pg/.s.PGSQL.5432"?
psql is the PostgreSQL interactive terminal.
What you are trying to do is connect to database without PostgreSQL running in first place.
Follow these steps:
Initialize the database with initdb -D .data
Start a PostgreSQL server with pg_ctl -D .data -l logfile start
Make sure it's running pg_ctl -D .data status
Connect to database with psql -d postgres (by default, postgres database is created)
At the end, make sure to stop database with pg_ctl -D .data stop when leaving the nix shell.
Assume we need to run second postgres instance in our production environment. First of them (postgres-one) already running and has few databases & data in them. Now I want to update my docker-compose.yaml file and add configuration for the second (postgres-two). Moreover I have to grab some databases info from postgres-one and copy it to postgres-two. Here how I'm trying to achieve this:
docker-compose.yaml
postgres-two:
image: postgres:12.5
depends_on:
postgres-one:
condition: service_started
...
ports:
- "5433:5432"
command: bash -c "chmod +x /usr/local/bin/init.sh && /usr/local/bin/init.sh"
volumes:
- ./data/postgres-two/init-db/init.sh:/usr/local/bin/init.sh
init.sh
#!/bin/bash
# allows you to skip the password prompt for pg_dump
echo "postgres-one:5432:dbname1_one:dbuser1_one:dbpass1_one" > ~/.pgpass
echo "postgres-one:5432:dbname2_one:dbuser2_one:dbpass2_one" >> ~/.pgpass
chmod 600 ~/.pgpass
# gets the data from external database & copies it to internal
pg_dump -h postgres-one -U dbuser1_one dbname1_one | psql -h localhost -U dbuser1_two -d dbname1_two
pg_dump -h postgres-one -U dbuser2_one dbname2_one| psql -h localhost -U dbuser2_two -d dbname2_two
But when I run this I get the error:
psql: error: could not connect to server: Connection refused
Is the server running on host "localhost" (127.0.0.1) and accepting
TCP/IP connections on port 5432?
I already tried it w/o -h localhost before, it gave me this as I remember:
psql: error: could not connect to server: No such file or directory Is the server running locally and accepting connections on Unix domain socket "/var/run/postgresql/.s.PGSQL.5432"?
If I add find /var/run/postgresql/ -name .s.PGSQL.5432 in the beginning of the init.sh it'll show nothing. So, as I understand I can't proceed further with psql because postgres server is not running at the moment. And of course I can't run it with postgres / pg_ctl commands because they can't be executed by root:
"root" execution of the PostgreSQL server is not permitted.
And of course docker containers run as the root user by default, and if I change the user it also give me errors such as:
chmod: changing permissions of '/usr/local/bin/init.sh': Operation not permitted
Am I doing something wrong? Or maybe I can get the dumps and apply them in other ways.. somehow?
I think server is failing to start because of volume mapping,
You can refer to this yml,
https://github.com/khezen/compose-postgres/blob/master/docker-compose.yml
Nvm, I solved the problem by placing the init.sh script into the docker-entrypoint-initdb.d folder. But then I also had an access issue with pg_dump because for some reason .pgpass didn't work in my case. For those, who will face the same problem, I fixed it with PGPASSWORD env like that:
PGPASSWORD="dbpass1_one" pg_dump -h postgres-one -U dbuser1_one dbname1_one | psql -h localhost -U dbuser1_two -d dbname1_two
I am running psql from a Debian 8.7 terminal but cannot figure out why it needs the -h flag to execute. For example, if i run psql -U postgres on the terminal, I get the following error:
psql: could not connect to server: No such file or directory
Is the server running locally and accepting
connections on Unix domain socket "/var/run/postgresql/.s.PGSQL.5432"?
However, when I add the host flag, I am able to connect:
psql -U postgres -h localhost
psql (9.4.10, server 9.6.2)
WARNING: psql major version 9.4, server major version 9.6.
Some psql features might not work.
Type "help" for help.
postgres=#
What is the cause?
It looks like the client and the server have a different idea about the directory where UNIX sockets should be created.
While connected to the database, run
SHOW unix_socket_directories;
That will tell you in which directory (or directories) the UNIX socket can be found.
You can use the directory name with the -h option or set the environment variable PGHOST to it for a local connection.
For example, if the result you get is /tmp (the factory default), you can use
psql -h /tmp -U postgres
While trying to connect to postgres running locally on my workstation, I get:
$ sudo -u postgres psql -c "create role ..."
could not change directory to "/home/esauer/workspace/cfme"
psql: could not connect to server: No such file or directory
Is the server running locally and accepting
connections on Unix domain socket "/tmp/.s.PGSQL.5432"?
My postgres-server install creates sockets in /var/run/postgresql.
How do I get the client to look in the proper location?
Check the --host option with psql --help.
Then you can make it permanent by setting it in your .psqlrc user file.
In your case try:
psql -h /var/run/postgresql -d your_database
I am trying to install postgresql on ubuntu.
I followed the steps from http://hocuspokus.net/2008/05/install-postgresql-on-ubuntu-804/.
And on typing the command :
psql template1
I am getting the following error:
psql: could not connect to server: No such file or directory
Is the server running locally and accepting
connections on Unix domain socket "/var/run/postgresql/.s.PGSQL.5432"?
The problem for me was that I had previously installed version 9.1, and 9.1 was still hogging the default port 5432. I was able to find this with the command:
root#labs:/home/caleb# /etc/init.d/postgresql status
9.1/main (port 5432): down
9.2/main (port 5433): online
So I was able to see that my 9.2 database was running on port 5433. So to connect, I had to explicitly specify port 5433:
psql -p 5433
try the following
psql template0
what resolved this error for me was deleting a file called postmaster.pid in the postgres directory. please see my question/answer using the following link for step by step instructions. my issue was not related to file permissions:
psql: could not connect to server: No such file or directory (Mac OS X)
You can also get in the CLI via this command:
psql -U postgres -p 5432 -h localhost
This should solve the error,
make a symbolic link to the /tmp/.s.PGSQL.5432:
sudo ln -s /tmp/.s.PGSQL.5432 /var/run/postgresql/.s.PGSQL.5432
Thanks to this post