Thread: ubuntu 18: PostgreSQL does not start. how can I totally remove and reinstall it

Hi there

I made somehow a mess with my PostgreSQL installation an an ubuntu 18.4 with 
PostgreSQL 10.0

service postgresql status
● postgresql.service - PostgreSQL RDBMS
    Loaded: loaded (/lib/systemd/system/postgresql.service; enabled; vendor 
preset: enabled)
    Active: active (exited) since Sat 2021-01-23 21:21:13 CET; 2min 26s ago
   Process: 1853 ExecStart=/bin/true (code=exited, status=0/SUCCESS)
  Main PID: 1853 (code=exited, status=0/SUCCESS)

I would like to remove everything, and reinstall.

How can I do that?

It seems, that

apt remove --purge postgresql
apt remove --purge postgresql-client

I not enough, as I have the same situation afterwards.

Or is there a way that I can rebuild PostgreSQL datastructure?

thanks

Robert

On 1/23/21 2:36 PM, robert rottermann wrote:
> Hi there
>
> I made somehow a mess with my PostgreSQL installation an an ubuntu 18.4 
> with PostgreSQL 10.0
>
> service postgresql status
> ● postgresql.service - PostgreSQL RDBMS
>    Loaded: loaded (/lib/systemd/system/postgresql.service; enabled; vendor 
> preset: enabled)
>    Active: active (exited) since Sat 2021-01-23 21:21:13 CET; 2min 26s ago
>   Process: 1853 ExecStart=/bin/true (code=exited, status=0/SUCCESS)
>  Main PID: 1853 (code=exited, status=0/SUCCESS)

What's the error?

>
> I would like to remove everything, and reinstall.
>
> How can I do that?
>
> It seems, that
>
> apt remove --purge postgresql
> apt remove --purge postgresql-client
>
> I not enough, as I have the same situation afterwards.

What happens after you purge the packages.

>
> Or is there a way that I can rebuild PostgreSQL datastructure?

Yes, but you haven't given us nearly enough information.

-- 
Angular momentum makes the world go 'round.

On 1/23/21 12:36 PM, robert rottermann wrote:
> Hi there
> 
> I made somehow a mess with my PostgreSQL installation an an ubuntu 18.4 
> with PostgreSQL 10.0
> 
> service postgresql status
> ● postgresql.service - PostgreSQL RDBMS
>     Loaded: loaded (/lib/systemd/system/postgresql.service; enabled; 
> vendor preset: enabled)
>     Active: active (exited) since Sat 2021-01-23 21:21:13 CET; 2min 26s ago
>    Process: 1853 ExecStart=/bin/true (code=exited, status=0/SUCCESS)
>   Main PID: 1853 (code=exited, status=0/SUCCESS)

What if you try:

sudo service postgresql@10-main status

Also  what does:

pg_lsclusters

show?



> 
> I would like to remove everything, and reinstall.
> 
> How can I do that?
> 
> It seems, that
> 
> apt remove --purge postgresql
> apt remove --purge postgresql-client
> 
> I not enough, as I have the same situation afterwards.
> 
> Or is there a way that I can rebuild PostgreSQL datastructure?
> 
> thanks
> 
> Robert
> 
> 
> 


-- 
Adrian Klaver
adrian.klaver@aklaver.com

On 1/23/21 12:36 PM, robert rottermann wrote:

Hi there

I made somehow a mess with my PostgreSQL installation an an ubuntu 18.4 with PostgreSQL 10.0

service postgresql status
● postgresql.service - PostgreSQL RDBMS
    Loaded: loaded (/lib/systemd/system/postgresql.service; enabled; vendor preset: enabled)
    Active: active (exited) since Sat 2021-01-23 21:21:13 CET; 2min 26s ago
   Process: 1853 ExecStart=/bin/true (code=exited, status=0/SUCCESS)
  Main PID: 1853 (code=exited, status=0/SUCCESS)

What if you try:

sudo service postgresql@10-main status

Also  what does:

pg_lsclusters

show?

root@elfero-test:~/scripts# pg_lsclusters
Ver Cluster Port Status Owner    Data directory              Log file
10  main    5433 online postgres /var/lib/postgresql/10/main /var/log/postgresql/postgresql-10-main.log
root@elfero-test:~/scripts# service postgresql@10-main status
● postgresql@10-main.service - PostgreSQL Cluster 10-main
   Loaded: loaded (/lib/systemd/system/postgresql@.service; indirect; vendor preset: enabled)
   Active: active (running) since Sat 2021-01-23 22:11:40 CET; 9h ago
  Process: 1348 ExecStart=/usr/bin/pg_ctlcluster --skip-systemctl-redirect 10-main start (code=exited, status=0/SUCCESS)
 Main PID: 1460 (postgres)
    Tasks: 7 (limit: 4543)
   CGroup: /system.slice/system-postgresql.slice/postgresql@10-main.service
           ├─1460 /usr/lib/postgresql/10/bin/postgres -D /var/lib/postgresql/10/main -c config_file=/etc/postgresql/10/main/postgresql.conf
           ├─1471 postgres: 10/main: checkpointer process  
           ├─1472 postgres: 10/main: writer process  
           ├─1473 postgres: 10/main: wal writer process  
           ├─1474 postgres: 10/main: autovacuum launcher process  
           ├─1475 postgres: 10/main: stats collector process  
           └─1476 postgres: 10/main: bgworker: logical replication launcher  

Jan 23 22:11:38 elfero-test systemd[1]: Starting PostgreSQL Cluster 10-main...

Jan 23 22:11:40 elfero-test systemd[1]: Started PostgreSQL Cluster 10-main.

Now I can also run psql. (why I could not before, I do not know ..)
BUT still I can not access the db from python.
even a simple python script like:

Produces:

root@elfero-test:~/scripts# python test_elfero.py
Traceback (most recent call last):
  File "test_elfero.py", line 2, in <module>
    conn = psycopg2.connect(host="localhost", user="postgres", database="postgres")
  File "/usr/lib/python2.7/dist-packages/psycopg2/__init__.py", line 130, in connect
    conn = _connect(dsn, connection_factory=connection_factory, **kwasync)
psycopg2.OperationalError: 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?

Any help is much appreciated

Robert

On Sun, Jan 24, 2021 at 2:58 PM robert@redo2oo.ch <robert@redo2oo.ch> wrote:
>
> root@elfero-test:~/scripts# pg_lsclusters
> Ver Cluster Port Status Owner    Data directory              Log file
> 10  main    5433 online postgres /var/lib/postgresql/10/main /var/log/postgresql/postgresql-10-main.log
> [...]
> psycopg2.OperationalError: 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?

It looks like your instance is configured to listen on port 5433, not 5432.

thanks a lot.
why dos such stupidity not hurt. ?????

have a nice weekend
robert

On 24.01.21 08:04, Julien Rouhaud wrote:
> On Sun, Jan 24, 2021 at 2:58 PM robert@redo2oo.ch <robert@redo2oo.ch> wrote:
>> root@elfero-test:~/scripts# pg_lsclusters
>> Ver Cluster Port Status Owner    Data directory              Log file
>> 10  main    5433 online postgres /var/lib/postgresql/10/main /var/log/postgresql/postgresql-10-main.log
>> [...]
>> psycopg2.OperationalError: 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?
> It looks like your instance is configured to listen on port 5433, not 5432.

Hi Robert,

> On 24. Jan, 2021, at 08:13, robert rottermann <robert@redcor.ch> wrote:
>
> thanks a lot.
> why dos such stupidity not hurt. ?????

the human brain does not contain pain generating or transmitting nerve cells, that's why. :-P

Cheers,
Paul

ARC - your Apple Authorised Service partner	H.D. Saviolaan 8
	B-1700 Dilbeek
	Belgium
info@arcict.com	www.arcict.com
tel. : +32 (0)2 466 50 00	fax. : +32 (0)2 466 88 33

On 1/23/21 11:08 PM, Hemil Ruparel wrote:
> One small note. Aside from the fact that your server is running from a 
> different port, psycopg2 does not support authentication without 
> password afaik. It has bit me twice

You will need to explain that further as I have not seen that over many 
years using psycopg2.



-- 
Adrian Klaver
adrian.klaver@aklaver.com

On Sun, 24 Jan 2021 at 08:09, Hemil Ruparel <hemilruparel2002@gmail.com> wrote:
>
> One small note. Aside from the fact that your server is running from a different port, psycopg2 does not support
authenticationwithout password afaik. It has bit me twice
 

It is more likely that you have some misunderstanding around
PostgreSQL authentication. If you have an example we can look into it.

-- Daniele

On 1/23/21 11:08 PM, Hemil Ruparel wrote:
> One small note. Aside from the fact that your server is running from a 
> different port, psycopg2 does not support authentication without 
> password afaik. It has bit me twice

Some thinking on this made me realize where you could get into this 
situation. Namely if you are using psycopg2 compiled against an older 
version of libpq that you then use to connect to a newer version of the 
server that is using an authentication method that the libpq version 
does not know about. This however is a generic issue of mismatched 
libraries and not unique to psycopg2.



-- 
Adrian Klaver
adrian.klaver@aklaver.com