[Pgpool-general] problem connecting through pgpool
Peter Vallero
pjvallero at yahoo.com
Tue Sep 11 06:05:03 UTC 2007
I tried resynching the databases (by shutting down both, tarring up database on db01 and untarring it on db02; then I restarted each database and then pgpool. I'm not getting the previous error, but still can't connect. Now when I try to connect from db01 to pgpool (on db01) I'm getting:
2007-09-10 22:48:54 DEBUG: pid 23272: I am 23272 accept fd 5
2007-09-10 22:48:54 DEBUG: pid 23272: Protocol Major: 3 Minor: 0 database: jacked user: jacked
2007-09-10 22:48:54 DEBUG: pid 23272: new_connection: connecting 0 backend
2007-09-10 22:48:54 LOG: pid 23272: pool_do_auth: maybe protocol version mismatch (current version 3)
2007-09-10 22:48:54 DEBUG: pid 23272: pool_read_string: read all from pending data. po:1 len:139
And when I try to connect from db02 to pgpool (on db01), I get:
2007-09-10 22:51:47 DEBUG: pid 23283: I am 23283 accept fd 5
2007-09-10 22:51:47 DEBUG: pid 23283: Protocol Major: 1234 Minor: 5679 database: user:
2007-09-10 22:51:47 DEBUG: pid 23283: SSLRequest: sent N; retry startup
2007-09-10 22:51:47 DEBUG: pid 23283: Protocol Major: 3 Minor: 0 database: jacked user: jacked
2007-09-10 22:51:47 DEBUG: pid 23283: new_connection: connecting 0 backend
2007-09-10 22:51:47 LOG: pid 23283: pool_do_auth: maybe protocol version mismatch (current version 3)
2007-09-10 22:51:47 DEBUG: pid 23283: pool_read_string: read all from pending data. po:1 len:139
Does pgpool have a problem with SSL communication?
Peter...
----- Original Message ----
From: Tatsuo Ishii <ishii at sraoss.co.jp>
To: pjvallero at yahoo.com
Cc: pgpool-general at pgfoundry.org
Sent: Monday, September 10, 2007 10:47:09 PM
Subject: Re: [Pgpool-general] problem connecting through pgpool
Where is pgpool running on? db01 ,db02 or any other host?
--
Tatsuo Ishii
SRA OSS, Inc. Japan
> pgpool.conf
> listen_addresses = '*'
> port = 9999
> pcp_port = 9898
> socket_dir = '/tmp'
> pcp_socket_dir = '/tmp'
> backend_socket_dir = '/tmp'
> pcp_timeout = 10
> num_init_children = 32
> max_pool = 8
> child_life_time = 300
> connection_life_time = 0
> child_max_connections = 0
> logdir = '/usr/local/pgpool/logs'
> replication_mode = true
> replication_strict = true
> replication_timeout = 5000
> load_balance_mode = false
> replication_stop_on_mismatch = false
> reset_query_list = 'ABORT; RESET ALL; SET SESSION AUTHORIZATION DEFAULT'
> print_timestamp = true
> master_slave_mode = false
> connection_cache = true
> health_check_timeout = 20
> health_check_period = 0
> health_check_user = 'nobody'
> insert_lock = false
> ignore_leading_white_space = true
> log_statement = false
> log_connections = false
> log_hostname = true
> parallel_mode = false
> enable_query_cache = false
> pgpool2_hostname = ''
> system_db_hostname = 'localhost'
> system_db_port = 5432
> system_db_dbname = 'pgpool'
> system_db_schema = 'pgpool_catalog'
> system_db_user = 'pgpool'
> system_db_password = ''
> backend_hostname0 = 'db01'
> backend_port0 = 5432
> backend_weight0 = 1
> backend_hostname1 = 'db02'
> backend_port1 = 5432
> backend_weight1 = 1
> enable_pool_hba = true
>
> pg_hba.conf (this is the same on each server)
> # "local" is for Unix domain socket connections only
> local all all trust
> host jacked jacked 206.188.27.0/24 trust
>
> pool_hba.conf
> # "local" is for Unix domain socket connections only
> local all all trust
> host jacked jacked 206.188.27.0/24 trust
>
>
> I have verified that I can connect from db01 to db01 and db02, and from db02 to db01 and db02 using psql.
>
>
> I read another post indicating that if the databases are not completely in synch (meaning OIDs and everything), then that might cause this problem. To create db02, I took a backup of the database on db01 and performed a restore/recovery on db02. Could this be the problem?
>
>
> Thanks,
>
> Peter...
>
>
> ----- Original Message ----
> From: Tatsuo Ishii <ishii at sraoss.co.jp>
> To: pjvallero at yahoo.com
> Cc: pgpool-general at pgfoundry.org
> Sent: Monday, September 10, 2007 9:35:13 PM
> Subject: Re: [Pgpool-general] problem connecting through pgpool
>
>
> It indicates that backend 1 failed to connect becauase of an
> authentication error. Can you show me the host/ip where pgpool running
> on, pgpool.conf and pg_hba.conf for backend 1?
> --
> Tatsuo Ishii
> SRA OSS, Inc. Japan
>
> > Thanks... I synched up my pg_hba.conf between the two nodes, and those with pool_hba.conf, and I'm no longer getting that exception. But I'm still not able to connect; now I'm getting this in the log:
> >
> > 2007-09-10 20:56:25 DEBUG: pid 21956: I am PCP 21956
> > 2007-09-10 20:56:37 DEBUG: pid 21935: I am 21935 accept fd 5
> > 2007-09-10 20:56:37 DEBUG: pid 21935: Protocol Major: 1234 Minor: 5679 database: user:
> > 2007-09-10 20:56:37 DEBUG: pid 21935: SSLRequest: sent N; retry startup
> > 2007-09-10 20:56:37 DEBUG: pid 21935: Protocol Major: 3 Minor: 0 database: jacked user: jacked
> > 2007-09-10 20:56:37 DEBUG: pid 21935: new_connection: connecting 0 backend
> > 2007-09-10 20:56:37 DEBUG: pid 21935: new_connection: connecting 1 backend
> > 2007-09-10 20:56:37 ERROR: pid 21935: pool_read_kind: kind does not match between master(69) slot[1] (82)
> >
> > Thanks for you help.
> >
> > Peter Vallero
> >
> >
> > ----- Original Message ----
> > From: Tatsuo Ishii <ishii at sraoss.co.jp>
> > To: pjvallero at yahoo.com
> > Cc: pgpool-general at pgfoundry.org
> > Sent: Monday, September 10, 2007 2:17:45 AM
> > Subject: Re: [Pgpool-general] problem connecting through pgpool
> >
> >
> > My guess is, you have trust authentication(i.e. no password) for
> > backend 0, while backend 1 uses md5 or some other password auth
> > methods. You need to use same auth method for all DBs and the auth
> > method must be one of trust, clear text password and pam.
> > --
> > Tatsuo Ishii
> > SRA OSS, Inc. Japan
> >
> > > I have configured pgpool-II in replication mode, and when I attempt to connect through pgpool to either of the 2 databases, I get this error:
> > >
> > > 2007-09-10 00:24:46 DEBUG: pid 8204: new_connection: connecting 0 backend
> > > 2007-09-10 00:24:46 DEBUG: pid 8204: new_connection: connecting 1 backend
> > > 2007-09-10 00:24:46 DEBUG: pid 8204: pool_read_message_length: slot: 0 length: 12
> > > 2007-09-10 00:24:46 DEBUG: pid 8204: pool_read_message_length: slot: 1 length: 8
> > > 2007-09-10 00:24:46 ERROR: pid 8204: pool_read_message_length: message length (8) in slot 1 does not match with slot 0
> > >
> > > Can anyone explain what is happening and give any advice to resolve this?
> > >
> > > Thanks,
> > >
> > > Peter Vallero
> > > Database Administrator
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://pgfoundry.org/pipermail/pgpool-general/attachments/20070910/e4dd6c85/attachment-0001.html
More information about the Pgpool-general
mailing list