Re: Postgresql 9.3 Server will not start after Ubuntu Upgrade

I am working on two systems, one running in Oracle VirtualBox on my
laptop, the other in a DigitalOcean droplet. I know on one of them I
tried to remove the postgres-9.6, and it must have been my laptop, here
is the output from pg_lsclusters:

Ver Cluster Port Status Owner    Data directory               Log file
9.3 main    5432 down   postgres /var/lib/postgresql/9.3/main
/var/log/postgresql/postgresql-9.3-main.log

Is it possible my effort to remove 9.6 was not complete or otherwise
screwed things up? (I dont remember the commands I used to do that, sorry)

Ken Beck
Liberty, Utah, USA

On 03/27/2018 12:17 PM, Adrian Klaver wrote:
> On 03/27/2018 11:00 AM, Ken Beck wrote:
>> I recently upgraded my OS from Ubuntu 14.04 LTS to 16.04 LTS and since,
>> postgresql refuses to re-start. I am concerned that if I poke around too
>> much, I may lose my old data. I have consulted various web postings here
>> and in ubuntu forums, and have not found an obvious solution. But please
>> excuse me, I am not a Linux expert, and I had a friend help me with
>> setup issues several years ago. They are no longer available, and I am
>> having trouble getting things to run.
>
> At a guess when you did the dist-upgrade(14.04 --> 16.04) you got the
> default version of Postgres for 16.04(9.6)
>
> Try, at the command line:
>
> pg_lsclusters
>
>
>>
>> Again, this is after upgrading from ubuntu 14.04 LTS to 16.04 LTS.
>> Consulting the book "PostgreSQL: Up and Running", I have looked at my
>> .conf files in /etc/postgresql/9.3/main, things seem to be in order.
>>
>> Here are the un-commented lines of postgresql.conf: (I have NOT changed
>> these, they are the settings that used to work in ubuntu 14.04)
>>
>> NOTE: my comments have // at beginning of the line, these lines are NOT
>> in the .conf file
>>
>> data_directory = '/var/lib/postgresql/9.3/main'        # use data in
>> another directory
>> hba_file = '/etc/postgresql/9.3/main/pg_hba.conf'    # host-based
>> authentication file
>> ident_file = '/etc/postgresql/9.3/main/pg_ident.conf'    # ident
>> configuration file
>>
>> external_pid_file = '/var/run/postgresql/9.3-main.pid'            #
>> write an extra PID file
>>
>> listen_addresses = 'localhost'        # what IP address(es) to listen
>> on;
>> port = 5432                # (change requires restart)
>> max_connections = 100            # (change requires restart)
>> unix_socket_directories = '/var/run/postgresql'    # comma-separated
>> list of directories
>> ssl = false                # (change requires restart)
>> ssl_cert_file = '/etc/ssl/certs/ssl-cert-snakeoil.pem'        # (change
>> requires restart)
>> ssl_key_file = '/etc/ssl/private/ssl-cert-snakeoil.key'        # (change
>> requires restart)
>>
>> // In the logging section, these are the only entries (is this why I
>> cant see any log files when I try to start the server??)
>> log_line_prefix = '%t '            # special values:
>> log_timezone = 'Navajo'
>>
>> // In Client Connections section, these are the only items active:
>> datestyle = 'iso, mdy'
>> timezone = 'Navajo'
>> lc_messages = 'en_US.UTF-8'            # locale for system error message
>>                      # strings
>> lc_monetary = 'en_US.UTF-8'            # locale for monetary formatting
>> lc_numeric = 'en_US.UTF-8'            # locale for number formatting
>> lc_time = 'en_US.UTF-8'                # locale for time formatting
>> default_text_search_config = 'pg_catalog.english'
>>
>> The pg_ctl.conf file is basically empty
>>
>> The pg_hba.conf file has these active lines:
>> local   all             postgres                                trust
>> local   all             all                                     peer
>> host    all             all             127.0.0.1/32            trust
>> host    all             all             ::1/128                 md5
>>
>> The start.conf file has one active line, 'auto'
>>
>> The pg_ident.conf file has no entries.
>>
>> When I try to start postgresql service using this line:
>> service postgresql reload
>>
>> then check running services using this:
>> systemctl list-units --type service --all
>> I see the lines related to postgres as follows:
>> postgresql.service                 loaded    active   exited  PostgreSQL
>> RDBMS
>> ● postgresql(at)9(dot)3-main(dot)service        loaded    failed   failed
>> PostgreSQL Cluster 9.3-main
>>
>> And, looking for log files, I find none.
>>
>> If I try to start pgAdmin, I see two servers on localhost, when I tried
>> to connect, I get an error first that says it can not open the log file,
>> on clicking that I get an error can't open file /home/XXX/.pgpass,
>> permission denied. Then I get a prompt for the password for postgres,
>> and when I enter the known password for the user, I then get a big 'Guru
>> Error' box with a message:
>> could not connect to server: Connection refused Is the server running on
>> host "127.0.0.1" and accepting TCP/IP connections on port 5432?
>> Note that the postgresql.conf file does list 5432 as the port number.
>>
>> I hope someone has the patience to look this over and suggest
>> remedies. --
>>
>> Ken Beck
>> Liberty, Utah, USA
>>
>>
>
>