[Pgpool-general] replication issue, only backend 0 is active

[Asaf Ohaion]

Using 2.3.3

I suspect the log line :
"new_connection: skipping slot 1 because backend_status = 3"

when asking pgpool to log all statements, and backends receiving
them, I can see that statement are only sent to backend 0

I can see that also in postgres logs,
backend 1 doesn't even show connection attempts

LOG:

2010-05-11 06:52:12 DEBUG: pid 4348: I am 4348
2010-05-11 06:52:12 DEBUG: pid 4349: I am 4349
2010-05-11 06:52:12 DEBUG: pid 4350: I am 4350
2010-05-11 06:52:12 DEBUG: pid 4351: I am 4351
2010-05-11 06:52:12 DEBUG: pid 4352: I am 4352
2010-05-11 06:52:12 DEBUG: pid 4353: I am 4353
2010-05-11 06:52:12 DEBUG: pid 4354: I am 4354
2010-05-11 06:52:12 DEBUG: pid 4355: I am 4355
2010-05-11 06:52:12 LOG:   pid 4281: pgpool-II successfully started.
version 2.3.2.2 (tomiteboshi)
2010-05-11 06:52:12 DEBUG: pid 4356: I am PCP 4356

2010-05-11 06:52:26 DEBUG: pid 4355: I am 4355 accept fd 5
2010-05-11 06:52:26 DEBUG: pid 4355: Protocol Major: 3 Minor: 0 database:
postgres user: postgres
2010-05-11 06:52:26 DEBUG: pid 4355: new_connection: connecting 0 backend
2010-05-11 06:52:26 DEBUG: pid 4355: new_connection: connecting 1 backend
2010-05-11 06:52:26 DEBUG: pid 4355: new_connection: skipping slot 1
because backend_status = 3
2010-05-11 06:52:26 DEBUG: pid 4355: pool_ssl: SSL requested but SSL
support is not available
2010-05-11 06:52:26 LOG:   pid 4355: pool_do_auth: maybe protocol version
mismatch (current version 3)
2010-05-11 06:52:26 DEBUG: pid 4355: pool_read_string: read all from
pending data. po:1 len:93

CONF:

# Host name or IP address to listen on: '*' for all, '' for no TCP/IP #
connections
listen_addresses = '*'

# Port number for pgpool
port = 5432

# Port number for pgpool communication manager
pcp_port = 9898

# Unix domain socket path.  (The Debian package defaults to
# /var/run/postgresql.)
socket_dir = '/tmp'

# Unix domain socket path for pgpool communication manager.
# (Debian package defaults to /var/run/postgresql)
pcp_socket_dir = '/tmp'


...default from sample..

# Replication mode
replication_mode = true

# Load balancing mode, i.e., all SELECTs are load balanced.
# This is ignored if replication_mode is false.
load_balance_mode = false

# if there's a data mismatch between master and secondary
# start degeneration to stop replication mode
replication_stop_on_mismatch = false

# If true, replicate SELECT statement when load balancing is disabled. #
If false, it is only sent to the master node.
replicate_select = false

..default..

# system DB info
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_hostname, backend_port, backend_weight
# here are examples
backend_hostname0 = 'localhost'
backend_port0 = 9999
backend_weight0 = 1
backend_data_directory0 = '/var/lib/pgsql/data'
backend_hostname1 = 'the.other.server.ip'
backend_port1 = 5432
backend_weight1 = 1
backend_data_directory1 = '/var/lib/pgsql/data'