[Pgpool-general] Error with streaming/balancing / pgpool 3.0.4

[Olivier NOEL]

 Hello,

I have an error with pgpool 3.0.4 in a streaming/balancing
configuration with 1 master and 2 slaves :

pgpool: 2011-08-16 10:51:58 ERROR: pid 1239: s_do_auth: unknown
response "E" before processing BackendKeyData
pgpool: 2011-08-16 10:51:58 ERROR: pid 1239: s_do_auth: unknown response "
pgpool: 2011-08-16 10:51:58 ERROR: pid 1239: s_do_auth: unknown response "
pgpool: 2011-08-16 10:51:58 ERROR: pid 1239: s_do_auth: unknown response "
pgpool: 2011-08-16 10:51:58 ERROR: pid 1239: s_do_auth: unknown
response "]" before processing BackendKeyData
pgpool: 2011-08-16 10:51:58 ERROR: pid 1239: pool_read2: failed to realloc

I followed the documentation on the website for the configuration of
the postgresql servers and pgpool.

DB servers are FreeBSD with latest Postgresql server, configured in
streaming mode (1 master and 2 slaves), replication/failover is
functional.

The pgpool host is a Debian Wheezy with pgpooladmin. Pgpool has been
install from packages. I have the same error without pgpooladmin.

I tested installing pgpool from sources and I have the same error.

I also tested pgpool on a FreeBSD box from ports and from sources, and
same error.

here is the config file :

listen_addresses = '*'

port = 5432

pcp_port = 9898

socket_dir = '/tmp'

pcp_socket_dir = '/tmp'

backend_socket_dir = '/tmp'

pcp_timeout = 10

num_init_children = 32

max_pool = 4

child_life_time = 300

connection_life_time = 0

child_max_connections = 0

client_idle_limit = 0

authentication_timeout = 60

logdir = '/tmp'

pid_file_name = '/tmp/pgpool.pid'

replication_mode = false

load_balance_mode = true

replication_stop_on_mismatch = false

failover_if_affected_tuples_mismatch = false

replicate_select = false

reset_query_list = 'ABORT; DISCARD ALL'

white_function_list = ''
black_function_list = 'nextval,setval'

print_timestamp = true

master_slave_mode = true

master_slave_sub_mode = 'stream'

delay_threshold = 10000000

log_standby_delay = 'if_over_threshold'

connection_cache = true

health_check_timeout = 10

health_check_period = 20

health_check_user = 'postgres'

failover_command = '/usr/local/etc/failover.sh %d "%h" %p %D %m %M "%H" %P'

failback_command = ''

fail_over_on_backend_error = true

insert_lock = false

ignore_leading_white_space = true

log_statement = false

log_per_node_statement = false

log_connections = false

log_hostname = false

parallel_mode = false

enable_query_cache = false

pgpool2_hostname = 'pglb01'

system_db_hostname = 'fpg01'
system_db_port = 5432
system_db_dbname = 'pgpool'
system_db_schema = 'pgpool_catalog'
system_db_user = 'pgpool'
system_db_password = 'password'


nable_pool_hba = false

recovery_user = 'postgres'

recovery_password = ''

recovery_1st_stage_command = 'basebackup.sh'

recovery_2nd_stage_command = ''

recovery_timeout = 90

client_idle_limit_in_recovery = 0

lobj_lock_table = ''

ssl = false

debug_level = 0
replication_timeout = 5000
backend_hostname0 = 'fpg01'
backend_port0 = 5432
backend_data_directory0 = '/usr/local/pgsql/data'
backend_weight0 = 1
backend_hostname1 = 'fpg02'
backend_port1 = 5432
backend_data_directory1 = '/usr/local/pgsql/data'
backend_weight1 = 2
backend_hostname2 = 'fpg03'
backend_port2 = 5432
backend_data_directory2 = '/usr/local/pgsql/data'
backend_weight2 = 2

-- 
Olivier