I need to write a script that will drop a PostgreSQL database. There may be a lot of connections to it, but the script should ignore that.
The standard DROP DATABASE db_name
query doesn't work when there are open connections.
How can I solve the problem?
This will drop existing connections except for yours:
Query pg_stat_activity
and get the pid values you want to kill, then issue SELECT pg_terminate_backend(pid int)
to them.
PostgreSQL 9.2 and above:
SELECT pg_terminate_backend(pg_stat_activity.pid)
FROM pg_stat_activity
WHERE pg_stat_activity.datname = 'TARGET_DB' -- ← change this to your DB
AND pid <> pg_backend_pid();
PostgreSQL 9.1 and below:
SELECT pg_terminate_backend(pg_stat_activity.procpid)
FROM pg_stat_activity
WHERE pg_stat_activity.datname = 'TARGET_DB' -- ← change this to your DB
AND procpid <> pg_backend_pid();
Once you disconnect everyone you will have to disconnect and issue the DROP DATABASE command from a connection from another database aka not the one your trying to drop.
Note the renaming of the procpid
column to pid
. See this mailing list thread.
In PostgreSQL 9.2 and above, to disconnect everything except your session from the database you are connected to:
SELECT pg_terminate_backend(pg_stat_activity.pid)
FROM pg_stat_activity
WHERE datname = current_database()
AND pid <> pg_backend_pid();
In older versions it's the same, just change pid
to procpid
. To disconnect from a different database just change current_database()
to the name of the database you want to disconnect users from.
You may want to REVOKE
the CONNECT
right from users of the database before disconnecting users, otherwise users will just keep on reconnecting and you'll never get the chance to drop the DB. See this comment and the question it's associated with, How do I detach all other users from the database.
If you just want to disconnect idle users, see this question.
PostgreSQL 13 introduced FORCE
option.
DROP DATABASE DROP DATABASE drops a database ... Also, if anyone else is connected to the target database, this command will fail unless you use the FORCE option described below. FORCE Attempt to terminate all existing connections to the target database. It doesn't terminate if prepared transactions, active logical replication slots or subscriptions are present in the target database.
DROP DATABASE db_name WITH (FORCE);
You could kill all connections before dropping the database using the pg_terminate_backend(int)
function.
You can get all running backends using the system view pg_stat_activity
I'm not entirely sure, but the following would probably kill all sessions:
select pg_terminate_backend(procpid)
from pg_stat_activity
where datname = 'doomed_database'
Of course you may not be connected yourself to that database
Easy Peasy.
I just restart the service in Ubuntu to disconnect connected clients.
sudo service postgresql stop
sudo service postgresql start
psql
DROP DATABASE DB_NAME;
dropdb DB_NAME
No need to go into postgres.
dropdb mydb --force
when there are active users.
Depending on your version of postgresql you might run into a bug, that makes pg_stat_activity
to omit active connections from dropped users. These connections are also not shown inside pgAdminIII.
If you are doing automatic testing (in which you also create users) this might be a probable scenario.
In this case you need to revert to queries like:
SELECT pg_terminate_backend(procpid)
FROM pg_stat_get_activity(NULL::integer)
WHERE datid=(SELECT oid from pg_database where datname = 'your_database');
NOTE: In 9.2+ you'll have change procpid
to pid
.
procpid
to pid
this snippet works on 9.3.
I noticed that postgres 9.2 now calls the column pid rather than procpid.
I tend to call it from the shell:
#!/usr/bin/env bash
# kill all connections to the postgres server
if [ -n "$1" ] ; then
where="where pg_stat_activity.datname = '$1'"
echo "killing all connections to database '$1'"
else
echo "killing all connections to database"
fi
cat <<-EOF | psql -U postgres -d postgres
SELECT pg_terminate_backend(pg_stat_activity.pid)
FROM pg_stat_activity
${where}
EOF
Hope that is helpful. Thanks to @JustBob for the sql.
PostgreSQL 9.2 and above:
SELECT pg_terminate_backend(pid)FROM pg_stat_activity WHERE datname = 'YOUR_DATABASE_NAME_HERE'
Here's my hack... =D
# Make sure no one can connect to this database except you!
sudo -u postgres /usr/pgsql-9.4/bin/psql -c "UPDATE pg_database SET datallowconn=false WHERE datname='<DATABASE_NAME>';"
# Drop all existing connections except for yours!
sudo -u postgres /usr/pgsql-9.4/bin/psql -c "SELECT pg_terminate_backend(pg_stat_activity.pid) FROM pg_stat_activity WHERE pg_stat_activity.datname = '<DATABASE_NAME>' AND pid <> pg_backend_pid();"
# Drop database! =D
sudo -u postgres /usr/pgsql-9.4/bin/psql -c "DROP DATABASE <DATABASE_NAME>;"
I put this answer because include a command (above) to block new connections and because any attempt with the command...
REVOKE CONNECT ON DATABASE <DATABASE_NAME> FROM PUBLIC, <USERS_ETC>;
... do not works to block new connections!
Thanks to @araqnid @GoatWalker ! =D
https://stackoverflow.com/a/3185413/3223785
In Linux command Prompt, I would first stop all postgresql processes that are running by tying this command sudo /etc/init.d/postgresql restart
type the command bg to check if other postgresql processes are still running
then followed by dropdb dbname to drop the database
sudo /etc/init.d/postgresql restart
bg
dropdb dbname
This works for me on linux command prompt
In my case i had to execute a command to drop all connections including my active administrator connection
SELECT pg_terminate_backend(pg_stat_activity.pid)
FROM pg_stat_activity
WHERE datname = current_database()
which terminated all connections and show me a fatal ''error'' message :
FATAL: terminating connection due to administrator command SQL state: 57P01
After that it was possible to drop the database
Nothing worked for me except, I loggined using pgAdmin4 and on the Dashboard I disconnected all connections except pgAdmin4 and then was able to rename by right lick on the database and properties and typed new name.
Success story sharing
; drop database TARGET_DB;
worked well in my case to make sure the db was gone by the time things started retrying.dropdb --force
.